Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
In January 2010 two groups independently published the observation that the depletion of CD8+ cells in SIV-infected macaques had no detectable impact on the lifespan of productively infected cells . This unexpected observation led the authors to suggest that CD8+ T cells control SIV viraemia via non-lytic mechanisms . However , a number of alternative plausible explanations , compatible with a lytic model of CD8+ T cell control , were proposed . This left the field with no consensus on how to interpret these experiments and no clear indication whether CD8+ T cells operated primarily via a lytic or a non-lytic mechanism . The aim of this work was to investigate why CD8+ T cells do not appear to reduce the lifespan of SIV-infected cells in vivo .
One of the most direct and convincing demonstrations of the importance of CD8+ T cells in controlling SIV infection in vivo is the observation that , on depleting CD8+ cells , SIV-1 viral load increases by 0 . 5–1 log . This observation has been made in both acute and chronic infection and has been replicated by a number of groups [1]–[3] . In 2010 two ground-breaking papers by Klatt et al and Wong et al , reported that following CD8+ cell depletion , although there was a robust increase in viral load , there was no increase in the lifespan of cells productively infected with SIV [4] , [5] . That is , when the lifespan of productively infected cells was measured there was no detectable difference between control macaques with an intact CD8+ T cell response and CD8+-depleted macaques . This highly unexpected result forced a re-evaluation of the role of CD8+ T cells in SIV infection and the authors concluded that CD8+ T cells controlled viral load ( since CD8+ cell depletion lead to an increase in viral load ) but that control was primarily via non-lytic mechanisms ( since CD8+ depletion did not increase the lifespan of infected cells ) . However , it has been argued that the data are not incompatible with a lytic mechanism of CD8+ T cell control . It has been hypothesised that ( i ) ART-treatment may impair CD8+ T cell killing ( ii ) CD8+ T cell killing may occur just before the cell would die anyway [6] ( iii ) CD8+ T cell killing may occur prior to viral production [4] , [7] or ( iv ) the measurements of lifespan may not be sufficiently accurate to detect a difference between depleted and control animals . The aim of this project was to firstly investigate whether the lack of an effect of CD8-depletion on lifespan is compatible with CD8+ T cell control of SIV viraemia via a lytic mechanism and then to establish whether the data best supports a lytic or non-lytic mechanism of CD8+ T cell control .
There are a number of explanations that potentially reconcile the lack of an effect of CD8-depletion on infected cell lifespan with a major cytolytic role for CD8+ T cells . We investigated each of these possibilities in turn . Having established that the estimated rates of viral clearance were not incompatible with purely lytic models of CD8+ T cell control we then investigated whether the dynamics of infection were more consistent with CD8+ T cells exerting their antiviral effects via a lytic or a non-lytic mechanism . We considered four models of lytic control and four models of non-lytic control . The lytic control models were chosen to describe the scenarios which ( in part 1 above ) were found to be qualitatively consistent with the lack of an effect of depletion on infected cell lifespan . The 4 lytic control models i ) a basic , widely used [15] , [16] , model of lytic control ii ) an extension of the basic model to include two populations of productively infected cells iii ) a model following Klenerman et al [6] in which SIV is cytopathic and iv ) a model following Althaus et al [7] in which CD8+ T cell killing is limited to the early non-productive stage of the viral lifecycle . The 4 non-lytic control models were: i ) a model in which non-lytic factors reduced new infection events e . g . beta-chemokines [17] , [18] ii ) an extension of model i to include two populations of productively infected cells iii ) a model in which non-lytic factors reduce virion production [19] iv ) an extension of model iii to include two populations of productively infected cells . To assess which of these models best describes infection dynamics we fitted each model to the time course of viral load and CD4+ T cell count in each macaque ( an average of 43 time points over 224 days ) and calculated the small sample Akaike's information criterion ( AICC ) of the best fitting parameter combination ( Figure 4 and 5 , and Figure S6 and Table S3 in Text S1 ) . Although the quality of the fits is rather poor , our results show clear and consistent support for the non-lytic model in which CD8+ T cells reduce infection; none of the lytic models receive any support for most data sets . The best-fitting non-lytic model ( non-lytic model ii , in which non-lytic factors reduced new infection events and there are two populations of productively infected cells ) was compared with each of the lytic models in turn . In every case the non-lytic model provided a significantly better fit ( lytic model i higher AICc 6/7 cases P = 0 . 043 , mean difference in AICc = 15; lytic model ii higher AICc 6/7 cases P = 0 . 028 , mean difference in AICc = 26; lytic model iii higher AICc 7/7 cases P = 0 . 018 , mean difference in AICc = 40; lytic model iv higher AICc 7/7 cases P = 0 . 018 , mean difference in AICc = 39 . All P values 2 tailed paired Mann-Whitney ) . Furthermore the non-lytic model in which infection was reduced performed consistently better than the non-lytic model in which virion production was reduced though the differences in performance were not as large as for the comparison between lytic and non-lytic models ( non-lytic model in which virion production was reduced ( iv ) had a higher AICc in 7/7 cases compared to the similar non-lytic model ii; P = 0 . 018 , 2 tailed paired Mann-Whitney , mean difference in the AICc = 10 ) . The reason why the lytic models fail can be seen from studying the equations . In the lytic models the rate of post-ART decline in viral load is determined by infected cell death due to viral toxicity and CD8+ T cell killing . Under the CD8 depletion regime post-ART decline in viral load is solely determined by infected cell death due to viral toxicity . Thus , to predict the similar post-ART decline under the two treatment-regimes lytic models need to attribute a small role to CD8+ cells . This small role of CD8+ T cells is poorly compatible with the increase in viral load following CD8 depletion in the absence of ART . Therefore , lytic models cannot accurately fit both post-ART decline and the increase in viral load upon depletion whereas non-lytic models can . Consequently models with a non-lytic component are likely to consistently outperform similar models with a lytic component . We conclude that although self-consistent hypotheses can be constructed in which CD8+ T cells exert their antiviral effects by lysis without a detectable impact on infected cell lifespan , these models are poorly predictive and a non-lytic model provides a better explanation of the viral load and CD4+ T cell dynamics .
SIV-infection of rhesus macaques is one of the best and most widely used animal models of HIV-1 infection . Indeed , the observation that CD8+ cell depletion in rhesus macaques causes an increase in viral load is frequently cited as strong evidence that CD8+ T cells control infection in HIV-1-infected humans . The finding that CD8+ T cell depletion has no discernable impact on infected cell lifespan in SIV-infected macaques is reminiscent of earlier observations in HIV-1-infected humans in which there was no relationship between infected cell lifespan and disease stage [6] . That is , in individuals with a high CD4+ T cell count and , presumably , a relatively intact CD8+ T cell response the lifespan of infected cells was very similar to individuals with a CD4 count <100 . The similarity of these observations underscores the importance and generality of the work of Klatt et al and Wong et al . We investigated a number of hypotheses to try and reconcile a major lytic role for CD8+ T cells with the observation that depletion of CD8+ T cells had little impact on the lifespan of infected cells . We found evidence that ART impairs CD8+ T cell function and allowing for this impairment increased the difference in lifespan between control and depleted animals but the difference remained small . Following Klenerman et al [6] and Althaus et al [7] we also investigated models in which CD8+ T cell killing occurred just prior to when a cell would die anyhow or prior to viral production . These models were qualitatively consistent but only for relatively narrow parameter ranges and required further assumptions ( namely that SIV is highly cytopathic or that infected cells are most vulnerable to CD8+ T cell killing prior to viral production , respectively ) . Unexpectedly , the simplest hypothesis: that the measurements of lifespan were not accurate enough to detect the difference between control and depleted animals did offer a plausible explanation . Depletion of CD8+ T cells results in a large increase in viral load so intuitively we expect that this implies that CD8+ T cells rapidly kill infected cells . If lysis is so rapid , it should clearly manifest as an increase in infected cell lifespan in the absence of CD8+ T cells . However , the large increase in viral load following depletion occurs over several days and , even if it is assumed to be due entirely to the removal of lytic CD8+ T cells , analysis shows that this only implies a rate of CD8+ T cell killing of 0 . 3 per day . That is , the rise in viraemia following CD8+ T cell depletion , indicates that CD8+ T cells are responsible for about 30% of productively infected cell death [20] . It is therefore not surprising that it is difficult to detect evidence for this relatively small contribution to cell death even in the extreme case of CD8-depleted macaques let alone in humans at different stages of infection . Next we investigated which mechanism provides the best explanation of SIV dynamics in acute and chronic infection . This analysis found strong support for non-lytic mechanisms of CD8+ T cell-mediated control of SIV infection , in particular via blocking of infection . The models we used were fully mechanistic , low parameter models which were constrained by CD8+ T cell data and were required to fit viral load and CD4+ T cell count data simultaneously over a wide dynamic range . Consequently the divergence between observation and prediction ( particularly for CD4+ T cell counts ) was rather large . Clearly , none of the models used represents the true model of SIV dynamics . However , the use of the AIC enables us to select the model which contains most information of the true model , i . e . is closest to reality [21] . Furthermore , the improved performance of the non-lytic models compared to the lytic models can be understood intuitively and is likely to be a general feature of comparisons between lytic and non-lytic models . We conclude that on their own , the lifespan estimates of Klatt and Wong cannot exclude the possibility that CD8+ T cell exert their antiviral effects via cytolysis . However , in conjunction with the failure of lytic models to compete with non-lytic models in predicting the time course of infection , the evidence favours a picture in which CD8+ T cells control infection via the production of non-lytic factors that reduce infection .
This study was carried out in strict accordance with the Association for Assessment and Accreditation of Laboratory Animal Care guidelines and the protocol was approved by the Emory University IACUC ( #062-2007Y ) . Ten SIVmac239-infected rhesus macaques were divided into two equal groups . Animals in group A were CD8+ lymphocyte depleted during early chronic phase . Animals in group B were CD8+ lymphocyte depleted during the late chronic phase . Antiretroviral therapy ( ART ) was given to all animals in both stages . Animals were given OKT8F ( CD8-depleting mAb ) for 3 consecutive days ( Group A , days 58–60 after infection; Group B , days 177–179 ) . ART ( PMPA and FTC ) was given for 28 consecutive days during both early and late chronic infection ( starting at d63 and d168 for group A , and d63 and d182 for group B ) . These studies were approved by the Emory University and University of Pennsylvania Institutional Animal Care and Use Committees . Further details are provided in [4] . Plasma viraemia was quantified by real-time reverse-transcriptase PCR . SIV-specific T cell responses were measured by intracellular cytokine staining for interferon-γ , tumour necrosis factor-α , and Interleukin-2 , as well as the degranulation marker CD107a , in response to pools of overlapping 15-mer peptides which spanned SIVmac239 gag , pol and env proteins [4] . The death rate ( 1/lifespan ) of productively infected cells was estimated by fitting Eq . 1 to viral load after initiation of ART [4] , [22] . ( 1 ) where A ( B ) is the contribution of short-lived ( long-lived ) productively infected cells respectively to viral load at time zero , µT ( µM ) is the death rate of short-lived ( long-lived ) productively infected cells ( d−1 ) respectively . The lytic model describes uninfected target cells ( T ) , productively infected cells ( T* ) and free virus ( V ) . Throughout , a dot denotes differentiation with respect to time . ( 2 ) ( 3 ) ( 4 ) Where λ is inflow of uninfected CD4+ cells ( cells . d−1 ) , β is infection rate , δT and δI death rate of uninfected and infected CD4+ cells respectively , k is CTL killing rate of infected CD4+ cells , p is production rate of free virions and c clearance rate of free virions ( all d−1 ) The fraction of CD8+ T cells , E , is entered as an empirical function obtained by linear interpolation between data points . With the exception of a few points , CD8+ and CD4+ T cell measurements were available whenever viral load measurements were made . An extension of the model with two populations of productively infected cells , T* and M* , with different death rates was also considered . This model is similar to the basic lytic model ( Eq . 2 to Eq . 4 ) but Eq . 2 and Eq . 3 are duplicated to describe , M* , the second population of infected cells . In Eq . 4 a term representing the production of virions by this population , pM* , is added . We assume that in vivo CTL killing follows the laws of mass action . The law of mass action has been shown to hold over a wide range of effector and target cell frequencies including the cell frequencies found in our experiments [23]–[26] . We estimated the decrease in infected cell death rate consistent with the increase in viral load following CD8+ T cell depletion if CD8+ T cells operate via a purely lytic mechanism . We used the basic lytic model ( Eq . 2 to Eq . 4 ) and , assuming constant uninfected target cells T , constant CD8+ T cell killing D prior to depletion , and a quasi-steady state between infected target cells and viral load we wrote Eq . 3 as: ( 5 ) Where b = βpT/c and D = kE . Prior to depletion dV/dt = 0 , so D = b-δI; after depletion D = 0 and the change in viral load was described by: ( 6 ) If the increase in viral load following CD8-depletion was solely due to the increase in lifespan of infected cells due to prevention of CD8 killing , then the estimate of slope ( b- δI ) gives an estimate of D . Alternative models including a model with target cell limitation and a model with an eclipse phase were also considered ( Text S1 ) . There are 4 methods that are widely used to estimate errors on parameter estimates in least-squares regression: the asymptotic covariance matrix method , bootstrapping the cases , bootstrapping the cases followed by trimming of extremes and bootstrapping the fit residuals [27]–[29] . No single method is optimal for all systems; the choice of method depends both on the model being fitted and the distribution of data . To assess the best approach for estimating the error on the death rate of short-lived productively infected cells in SIV-infected macaques we generated data in silico using the equation for viral decline after initiation of ART ( Eq . 1 ) , added random noise , fitted the model to the in silico “data” and then compared the accuracy of the 95% confidence intervals generated using the four methods . Further details are provided in Text S1 . The distribution of bootstrap sample estimators approximates the distribution of the parameter estimator [27] so we constructed the distribution of the death rate estimates in depleted and control animals by fitting Eq . 1 to the viral load data in the 10 macaques following initiation of ART and calculated 5000 trimmed bootstrap estimates of the death rate for each animal . From these distributions we sampled , with replacement , a death rate for each animal and calculated the mean for the five animals in the control and depleted group and then the difference in the mean between the two groups . We repeated this 1000 times . To investigate if impaired CD8+ T cell function after ART-treatment , represented by reduced CD107-expression , can explain the similar death rate estimates found in CD8-depleted and control animals we explicitly included CD107-expression into the lytic model with two populations of infected cells ( T* and M* ) . During ART treatment , infection rate β = 0 , so the model reduces to: ( 7 ) ( 8 ) ( 9 ) Where δI and δS are the death rates of the T* and M* population and pT and pM are production rates of free virions by the T* and M* population respectively ( all in d−1 ) . Fraction f is the proportion of infected cell death attributable to CD8+ T cell killing; ( 1-f ) is the proportion attributable to all other factors . Impaired CD8+ T cell function was represented by the fraction of CD107+ CD8+ T cells , E ( t ) , relative to the pre-ART fraction of CD107+ CD8+ T cells , E0 . E ( t ) was obtained by linear interpolation between data points . Resulting death rate estimates in control animals ( i . e . corrected for CD8+ T cell impairment ) were compared with death rates estimated in depleted animals using a Mann-Whitney test for different values of f . It has been suggested [6] that infected cells initially produce few virions while the bulk of virions are produced just before the cells dies . This was represented by 2 populations of SIV-infected cells: population L* consists of recently infected cells that do not produce virions and die at a negligible rate; population A* represents productively infected cells . The population of uninfected CD4+ T cells was described by Eq . 2 . ( 10 ) ( 11 ) ( 12 ) Where is transition rate from the L* to the A* population and δI and δA are the death rates of the L* to the A* population ( all d−1 ) . This is based on [7] and represented by two populations of SIV-infected cells: population I* consists of recently infected cells which are susceptible to CD8+ T cell-mediated killing , population P* consists of productively infected cells that evade CTL-killing through down-regulation of MHC-I molecules . Uninfected CD4+ T cells were described by Eq . 2 . ( 13 ) ( 14 ) ( 15 ) Where γ is the transition rate from the I* to the P* population and δP is the death rate of the P* population ( both d−1 ) . The non-lytic models share the basic description of T , T* and V dynamics ( Eq . 2 to Eq . 4 ) . However , in the non-lytic models CD8+ T cells do not kill infected cells ( k = 0 ) but decrease infection rate β or virion production rate p by a fraction : ( 16 ) The four different lytic and non-lytic models were fitted to data of virus load and total CD4+ T cell population , scaled to obtain equal mean . E ( t ) was obtained by linear interpolation of experimental data . ART-treatment was simulated by setting infection rate β = 0 . To fit the models to the data we used the pseudorandom algorithm in the modFit-function of the FME-package in R [30] . The small sample ( second-order ) bias adjusted Akaike Information Criterion ( AICc ) [21] , [31] was used to compare the fit of the models . The model with the lowest AICc is considered to describe the experimental data best . As a rule of thumb we assumed considerable support for models with an AICc within two of the lowest; if models differ by three to seven AICc units from the minimum AICc there is some support for the model with the higher AICc while a difference larger than 10 suggests that the model is very unlikely [21] . | Several studies have shown a role for CD8+ T cells in controlling SIV-infection . However , early last year two groups independently showed that depletion of CD8+ lymphocytes did not result in a measurable increase in the lifespan of productively infected cells , suggesting that direct cell killing may not be the major mechanism of antiviral activity by CD8+ lymphocytes . We investigated whether the lack of an effect on lifespan of infected cells indeed excludes a lytic role for CD8+ cells and whether a non-lytic effect of CD8+ cells , for instance by preventing new infections or blocking production of free virions , better explains the similar death rates of SIV-infected cells in animals with and without CD8+ lymphocytes . We found that , even though lytic models of CD8+ cell function are compatible with the absence of an effect of CD8+ cells on the lifespan of productively infected cells , the most likely mechanism of CD8-control in SIV-infection is via a non-lytic mechanism . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"immune",
"cells",
"theoretical",
"biology",
"hiv",
"t",
"cells",
"immunology",
"biology",
"viral",
"diseases",
"immune",
"response"
] | 2011 | Why Don't CD8+ T Cells Reduce the Lifespan of SIV-Infected Cells In Vivo? |
An estimated 50 million dengue virus ( DENV ) infections occur annually and more than forty percent of the human population is currently at risk of developing dengue fever ( DF ) or dengue hemorrhagic fever ( DHF ) . Despite the prevalence and potential severity of DF and DHF , there are no approved vaccines or antiviral therapeutics available . An improved understanding of DENV immune evasion is pivotal for the rational development of anti-DENV therapeutics . Antagonism of type I interferon ( IFN-I ) signaling is a crucial mechanism of DENV immune evasion . DENV NS5 protein inhibits IFN-I signaling by mediating proteasome-dependent STAT2 degradation . Only proteolytically-processed NS5 can efficiently mediate STAT2 degradation , though both unprocessed and processed NS5 bind STAT2 . Here we identify UBR4 , a 600-kDa member of the N-recognin family , as an interacting partner of DENV NS5 that preferentially binds to processed NS5 . Our results also demonstrate that DENV NS5 bridges STAT2 and UBR4 . Furthermore , we show that UBR4 promotes DENV-mediated STAT2 degradation , and most importantly , that UBR4 is necessary for efficient viral replication in IFN-I competent cells . Our data underscore the importance of NS5-mediated STAT2 degradation in DENV replication and identify UBR4 as a host protein that is specifically exploited by DENV to inhibit IFN-I signaling via STAT2 degradation .
Approximately fifty million dengue virus ( DENV ) infections occur annually with more than forty percent of the human population at risk of developing dengue fever ( DF ) or dengue hemorrhagic fever ( DHF ) . DF and its more severe form , DHF , are potentially fatal diseases caused by the four serotypes ( 1 , 2 , 3 and 4 ) of DENV [1] . As no approved vaccines or antiviral therapeutics are available for the prevention or treatment of DENV infections , it is imperative that the biology and immunology of DENV infections are better understood . An in depth comprehension of DENV-host interactions will accelerate our progress in developing DENV therapeutics . DENV , along with other clinically relevant arboviruses such as West Nile virus ( WNV ) , Japanese encephalitis virus ( JEV ) and yellow fever virus ( YFV ) , belongs to the flavivirus genus of the Flaviviridae family . The flavivirus genome is a capped 11 kb genome that is translated into a single polyprotein , which is cleaved both by the viral protease ( NS2B/3 ) and host proteases to yield three structural proteins ( capsid [C] , membrane [prM/M] and envelop [E] ) and seven non-structural proteins ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B and NS5 ) [2] , [3] . The flavivirus structural proteins incorporate the viral genome into newly generated virions while the non-structural proteins replicate the viral genome and exploit the cellular machinery to subvert host immune responses . The approximately 900-amino-acid NS5 protein is the largest and most conserved flavivirus protein [4] . This multifunctional protein has RNA-dependent RNA polymerase ( RdRp ) activity as well as methyltransferase activity [5] , [6] , [7] , [8] , [9] . In addition , more recent studies have shown that NS5 is a potent interferon-signaling antagonist [10] , [11] , [12] , [13] , [14] , [15] , [16] . The significance of the interferon ( IFN ) response as an important component of host immunity is underscored by numerous examples of viruses that antagonize it [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] . Viruses express pathogen-associated molecular patterns ( PAMPs ) that trigger the production of type I IFN ( IFNα/β or IFN-I ) [26] . Binding of IFN-I to the cell-surface IFN-I receptor ( IFNAR ) initiates a signaling cascade that results in the activation and phosphorylation of the Janus kinases , Jak1 and Tyk2 , and the transcription factors , STAT1 and STAT2 . Phosphorylated STAT1 and STAT2 along with IRF9 form the heterotrimeric transcriptional complex , ISGF3 [27] , [28] , and induce the expression of antiviral IFN-stimulated genes ( ISGs ) [29] , [30] , [31] , [32] . DENV encodes several antagonists of both IFN-I production and IFN-I signaling [13] , [14] , [33] , [34] , [35] , [36] , [37] , [38] . The NS5 proteins of DENV and other flaviviruses have been shown to be potent inhibitors of IFN signaling . NS5 proteins of different flaviviruses may target different steps of the IFN signaling pathway . For example , WNV NS5 prevents the phosphorylation of the STAT proteins , while DENV NS5 binds human STAT2 and promotes its proteasomal degradation [13] , [15] . Although STAT degradation is a common mechanism of viral IFN antagonism [19] , [20] , [39] , [40] , the requirements for DENV NS5-mediated STAT2 degradation are unique . DENV NS5-mediated STAT2 degradation requires NS5 to be proteolytically cleaved at its N terminus from a larger precursor protein [13] . N-terminal cleavage of NS5 normally occurs during DENV infection because the NS2B/3 protease cleaves at the junction located between NS4B and NS5 thereby releasing NS5 from the viral polyprotein [41] . Though both unprocessed and proteolytically-processed NS5 can bind STAT2 , only processed NS5 can efficiently mediate STAT2 degradation [13] . Furthermore , the first ten amino acids of NS5 are dispensable for STAT2 binding but are indispensable for STAT2 degradation [13] . While the viral requirements for DENV-mediated STAT2 degradation are known , the cellular components were unspecified until now . This study identifies the 600-kDa protein , UBR4 , as a binding partner of DENV NS5 . UBR4 is a member of the N-recognin family , which contains proven and predicted E3 ligases that recognize and degrade proteins containing destabilizing N termini [42] . UBR4 interacts preferentially with proteolytically-processed DENV NS5 but not with YFV NS5 or WNV NS5 , highlighting the specificity of the DENV NS5/UBR4 interaction . Furthermore , we have identified two residues within the first 10 N-terminal amino acids of NS5 , threonine 2 and glycine 3 , that are required for NS5 binding to UBR4 . These two residues are conserved across the four DENV serotypes but are not found in other flaviviruses . Finally , we show that UBR4 is required for DENV-mediated STAT2 degradation , and for efficient DENV replication in IFN-I competent cells . Our data confirm the importance of NS5-mediated STAT2 degradation for DENV replication , and identify UBR4 as a host protein that is specifically co-opted by DENV to inhibit IFN-I signaling via STAT2 degradation .
DENV NS5 binds human and non-human primate STAT2 but cannot efficiently mediate STAT2 degradation unless it is expressed in the context of a precursor protein from which it is N-terminally cleaved [13] . When DENV NS5 is engineered to be expressed downstream of ubiquitin , cellular hydrolases cleave ubiquitin in a manner that mimics the cleavage of NS4B away from NS5 by the NS2B/3 protease during DENV infection [13] . To identify host proteins that are required for NS5-mediated degradation of STAT2 , we generated a DENV2 NS5 construct consisting of RFP-ubiquitin fused to the NS5 N-terminus and a TAP ( tandem-affinity purification ) tag fused to the NS5 C-terminus . This NS5 construct was expressed in 293T cells , in the presence or absence of human STAT2-FLAG , and then purified using the TAP method . A high molecular weight protein band was consistently and specifically co-precipitated with NS5 both in the presence and absence of overexpressed STAT2 ( Figure 1A ) . Trypsin digestion of this band yielded five peptides that were identified by mass spectrometry as sequences of the N-recognin , UBR4 ( Table 1 ) [42] . To confirm the binding and specificity of the interaction between DENV2 NS5 and UBR4 , HA-tagged DENV2 NS5 , DENV1 NS5 , and YFV NS5 were expressed in 293T cells and purified by immunoprecipitation with antibodies raised against the HA epitope . The NS5 protein of both DENV1 and DENV2 precipitated UBR4 from 293T cells but YFV NS5 was unable to precipitate UBR4 from these cells ( Figure 1B ) . WNV NS5 was also unable to bind UBR4 ( Figure 1C ) . In order to assess the specificity of DENV NS5 for UBR4 , we also examined the ability of NS5 to bind another member of the N-recognin family , UBR5 . DENV1 and DENV2 NS5 bound UBR4 but not UBR5 , further highlighting the unique interaction between UBR4 and DENV NS5 ( Figure 1B ) . UBR4 was found to bind both processed ( RFP-ubiquitin-NS5-TAP ) ( Figure 1A ) , as well as unprocessed NS5 ( NS5-HA ) ( Figure 1B ) . Since cleavage of NS5 promotes STAT2 degradation , we tested whether proteolytic processing would also affect the binding efficiency of NS5 for UBR4 . The E domain in Figure 2A refers to the E protein of DENV2 . Inclusion of a protein ( such as E or RFP ) upstream of ubiquitin allows one to differentiate between cleaved and uncleaved NS5 [13] . HA-tagged unprocessed NS5 ( NS5-HA ) and processed NS5 ( proNS5-HA ) ( Figure 2A ) were expressed in 293T cells and the NS5 proteins were immunoprecipitated and tested for UBR4 binding . Although both constructs precipitated UBR4 , proNS5-HA precipitated two-fold higher amounts of UBR4 than NS5-HA did ( as quantified by densitometry ) ( Figure 2B ) . Given that the first 10 amino acids of NS5 are dispensable for STAT2 binding but indispensable for STAT2 degradation [13] , we asked whether this N terminal region of NS5 was also important for UBR4 binding . To test this , we expressed and immunoprecipitated processed HA-tagged DENV NS5 proteins containing a deletion of 10 or 306 residues at their N-termini , and assessed their ability to bind UBR4 . Full length HA-tagged NS5 ( NS5-HA ) was able to precipitate UBR4 and STAT2 , and its ability to precipitate UBR4 , but not STAT2 , was increased seven-fold ( as quantified by densitometry ) when DENV NS5 was proteolytically-processed ( proNS5-HA ) ( Figure 2C ) . Proteolytically-processed NS5 lacking the first 10 amino acids ( proΔ10NS5-HA ) precipitated STAT2 but not UBR4 , and proteolytically-processed NS5 lacking the first 306 amino acids ( proΔ306NS5-HA ) precipitated neither protein ( Figure 2C ) . When protein levels were examined in the whole cell extracts ( WCE ) , STAT2 was reduced in proNS5-HA-expressing cells and slightly reduced in NS5-HA-expressing cells compared with proΔ10NS5-HA- or proΔ306NS5-HA-expressing cells ( Figure 2C ) , which is consistent with published reports [13] . Thus , only the NS5 proteins that bound UBR4 could mediate STAT2 degradation , and increased UBR4 binding by NS5 correlated with increased NS5-mediated STAT2 degradation . The interaction of NS5 with UBR4 and the requirement for the first 10 amino acids of NS5 in mediating this DENV NS5-UBR4 interaction was also observed by NS5-UBR4 colocalization using immunofluorescence analysis in Vero cells ( Figure 2D ) . To further define which of the N-terminal residues of DENV NS5 are required for its interaction with UBR4 , alanine scanning of the first 5 amino acids of DENV NS5 was conducted ( Figure 2E ) . Immunoprecipitation experiments with these mutant proteins revealed that the threonine ( T ) at position 2 and the glycine ( G ) at position 3 , which are conserved among the DENV serotypes but absent in other flaviviruses , were required for NS5/UBR4 interaction and NS5-mediated STAT2 degradation ( Figure 2F ) . The fact that NS5 mutants lacking residues T2 or G3 bound STAT2 but not UBR4 ( Figure 2F ) suggested that the interaction between NS5 and UBR4 was independent of STAT2 . To confirm this result , the STAT2-deficient U6A cell line [43] was transfected with proNS5-HA or proΔ10NS5-HA . ProNS5-HA , but not proΔ10NS5-HA , precipitated UBR4 from U6A cells ( Figure 3A ) . Unlike with human STAT2 ( hSTAT2 ) , NS5 does not bind and subsequently degrade mouse STAT2 ( mSTAT2 ) [44] . The STAT2 proteins of mouse and human are divergent and share only 70% identity but the UBR4 proteins of mouse and human are 97% identical . When proNS5-HA was expressed in mouse cells ( Hepa1 . 6 ) , mouse UBR4 bound proNS5-HA but not proΔ10NS5-HA confirming that an interaction between NS5 and STAT2 is not required for NS5 to interact with UBR4 ( Figure 3A ) . These data suggest that NS5 requires binding to both UBR4 and STAT2 to mediate STAT2 degradation . NS5 binds the coiled-coil region located within the first half of hSTAT2 [44] . Though mSTAT2 and human STAT1 ( hSTAT1 ) cannot bind NS5 , chimeric proteins that replace the first 301 amino acids of mSTAT2 ( h/mSTAT2 ) or the first 316 amino acids of hSTAT1 ( hSTAT2/1 ) with those of hSTAT2 can bind NS5 [44] . We expressed and immunoprecipitated FLAG-tagged STAT proteins and STAT chimeric proteins in the presence or absence of proNS5-HA . When hSTAT2 was overexpressed , STAT2 degradation was not observed because STAT2 degradation was likely masked by the large amount of overexpressed STAT2 present ( Figure 3B ) . However , we observed that while hSTAT1 and mSTAT2 did not bind UBR4 , h/mSTAT2 , hSTAT2/1 and hSTAT2 all bound UBR4 in the presence of proNS5-HA ( Figure 3B ) . Since only those STAT molecules that could bind NS5 could also bind UBR4 , and NS5 binds UBR4 in the absence of STAT2 , we conclude that NS5 serves as a bridge molecule between STAT2 and UBR4 . These results were confirmed in the context of DENV infection of a transformed human cell line ( 293T ) and primary untransformed human cells ( monocyte-derived dendritic cells or MDDCs ) ( Figure 3C and 3D ) . Cells were infected with DENV2 at an MOI of 3 for 24 hours , and lysed for immunoprecipitation with STAT2 antibodies or control IgG . Although the majority of STAT2 was degraded during DENV infection , the remaining STAT2 co-immunoprecipitated UBR4 from DENV-infected cells but not from mock-infected cells ( Figure 3C and 3D ) , which is consistent with NS5 binding to and bringing together STAT2 and UBR4 during DENV infection . We next assessed the functional relevance of UBR4 in DENV-mediated STAT2 degradation . To test if UBR4 is required for DENV-mediated STAT2 degradation , UBR4 levels were stably reduced in 293T cells using small hairpin RNA ( shRNA ) directed against UBR4 . Three stable UBR4-knockdown cell lines were generated using shRNA that targeted different sequences within UBR4 . The cells were mock infected or infected with DENV2 at an MOI of 10 , and lysed for western blot analysis at 4 , 8 , 12 and 24 hours post-infection . When cells expressing control non-targeting shRNA were infected with DENV2 , STAT2 levels decreased by 4 hours post-infection . However , in the three independently-derived , UBR4-deficient 293T cell lines , STAT2 levels decreased at a slower rate ( Figure 4A ) . Furthermore , NS5 levels were lower in the UBR4-knockdown cells than in the control cells suggesting that there was a DENV replication defect in UBR4-knockdown cells . The similar phenotype of the three UBR4-knockdown cell lines and their difference from the control cell line indicated that the effect of UBR4 knockdown on DENV-mediated STAT2 degradation was due to the decreased level of UBR4 and not to off-target effects of the shRNA . We next examined the functional relevance of UBR4 in mediating STAT2 degradation with the other three DENV serotypes ( DENV1 , 3 or 4 ) . STAT2 levels were higher and NS5 levels were lower in shUBR4-expressing cells than in control cells ( Figure 4B ) . Thus , UBR4 is required for efficient STAT2 degradation mediated by all four DENV serotypes . The UBR4 gene is predicted to produce several splice variants encoding proteins of greater than 5000 amino acids . Since it is unclear which UBR4 isoform is required for DENV-mediated STAT2 degradation , we cloned a region of UBR4 ( UBR4-NT ) that is predicted to be present in all the large UBR4 isoforms and which also contains the UBR box , a 70-amino-acid zinc-finger-like domain required for recognition of N-end rule substrates [42] . The UBR box is located between amino acids 1662–1723 of the UBR4 reference sequence ( NCBI Accession # Q5T4S7 ) , and the UBR4-NT clone encodes amino acids 1–2233 of the reference sequence . Co-immunoprecipitation experiments revealed that proNS5-HA did not bind amino acids 1–2233 of UBR4 ( Figure 4C ) indicating that sequences in the C-terminal half of UBR4 are required to mediate its interaction with NS5 . Also , expression of UBR4-NT had no effect on DENV-mediated STAT2 degradation ( Figure 4D ) . The experiments in Figure 4 confirm that a functional UBR4-NS5-STAT2 complex is required for efficient STAT2 degradation and that multiple domains of UBR4 are required for this function . The ability of DENV to degrade STAT2 determines how well it replicates in an IFN-I-competent cell [44] . Thus , a protein that is required for DENV-mediated STAT2 degradation should also enhance DENV replication in IFN-I-competent cells . To test if UBR4 is required for DENV replication , UBR4-knockdown 293T cells were infected with DENV2 at multiplicities of infection ( MOI ) of 0 . 1 , 1 and 10 , and measured for virus at 24 hours post-infection . DENV replicated to lower levels in UBR4-knockdown cells than in control cells ( Figure 5A ) . The replication defect was most striking at a lower MOI and an approximately 10-fold decrease in virus levels was observed in shUBR4 cells with an MOI of 0 . 1 of DENV . In contrast , UBR4 depletion had no effect on the replication of YFV or encephalomyocarditis virus ( EMCV ) , a positive-strand RNA virus belonging to the Picornaviridae family , indicating a specific requirement of UBR4 in DENV replication ( Figure 5A ) . DENV1 , 3 and 4 also replicated to lower levels in UBR4-knockdown cells than in control cells indicating that UBR4 is required for the efficient replication of all four DENV serotypes ( Figure 5B ) . Since UBR4 was required for DENV-mediated STAT2 degradation , we hypothesized that the DENV replication defect in UBR4-deficient cells was due to an inability of DENV to antagonize IFN-I signaling by degrading STAT2 . If this is the case , lack of IFN-I should compensate for the requirement of UBR4 in DENV replication . To test this , we infected control and UBR4-knockdown Vero cells with DENV . Vero cells lack IFN-I genes and therefore cannot make IFN-I in response to viral infection [45] . DENV replicated to similar levels in UBR4-knockdown and control Vero cells ( Figure 5C ) . Yet when Vero cells were infected with DENV and then exogenously treated with IFN-I 6 hours later , a DENV replication defect was observed in the UBR4-deficient Vero cells ( Figure 5C ) . Protein levels of NS5 , UBR4 , and STAT2 in UBR4-knockdown Vero cells showed that UBR4 levels were indeed lower and that DENV-mediated STAT2 degradation was defective in UBR4-knockdown cells ( Figure 5B ) . Treating DENV-infected UBR4-knockdown 293T cells with a neutralizing anti-IFNAR antibody corroborated the effect of IFN-I on DENV replication in UBR4-deficient Vero cells ( Figure 5D ) . We observed a significant increase in DENV replication in UBR4-knockdown 293T cells treated with the neutralizing anti-IFNAR antibody compared to UBR4-knockdown 293T cells treated with IgG control antibodies . This contrasts with what was observed in control 293T cells where DENV replication was unaffected by treatment with anti-IFNAR antibodies ( Figure 5D ) . IFN exerts its biological effect by upregulating interferon-stimulated genes ( ISGs ) , which encode products that restrict viral replication . To examine the biological relevance of UBR4 in preventing the antiviral action of IFN-I during DENV infection , we examined the induction of ISG54 mRNA in UBR4-knockdown 293T cells . There was a significant induction of ISG54 mRNA in UBR4-knockdown cells during DENV infection compared to control cells ( Figure 5E ) . These results indicate that UBR4 is required for preventing the antiviral action of IFN-I during DENV infection . Dendritic cells are thought to be an important cell type in which DENV replicates in vivo [46] , [47] . We reduced UBR4 levels in primary MDDCs from five donors using shRNA lentiviral constructs , and tested the effect of this decrease on DENV replication . When MDDCs from each donor were infected at an MOI of 3 , approximately 35% of transfected cells were highly infected and showed viral glycoprotein ( E ) expression by FACS . At 12 hours post infection , as expected , the levels of UBR4 were decreased in the UBR4-knockdown cells compared to control cells ( Figure 6A ) . When the levels of ISG15 , RIG-I and ISG54 mRNA were analyzed in these cells , more ISGs were induced in four of the five donors ( Figure 6B , 6C and 6D respectively ) . In addition , more DENV was present in control cells than in UBR4-knockdown cells at 48 hours post infection ( Figure 6E ) . Thus , UBR4 is required for inhibiting ISG induction and increasing DENV replication in a primary cell type that is of importance in DENV infections .
The IFN-I response is one of the first lines of protection against DENV infection , and serves to curb viral replication and dissemination by generating an antiviral intracellular environment [48] . The potency of the type I IFN pathway is exemplified by the fact that DENV antagonizes both IFN synthesis and IFN signaling in order to ensure its replication and survival [13] , [14] , [33] , [34] , [35] , [36] , [37] , [38] . DENV NS5 inhibits IFN-I signaling by mediating proteasome-dependent STAT2 degradation , and STAT2 degradation promotes DENV replication [13] , [14] , [44] . With this study , we report the discovery of a host factor , UBR4 , that is essential for DENV-dependent STAT2 degradation . We describe the interaction of UBR4 with NS5 and show that this interaction is crucial for inhibiting type-I IFN signaling and promoting efficient DENV replication . UBR4 associates with DENV NS5 but not with the closely related YFV NS5 or WNV NS5 . UBR4 also binds preferentially to proteolytically-processed DENV NS5 , which is the form of NS5 that efficiently mediates STAT2 degradation . Binding of UBR4 to DENV NS5 requires amino acids T2 and G3 of NS5 , which are also critical for STAT2 degradation . These amino acids are conserved amongst the four DENV serotypes but are absent in other flaviviruses ( Figure 2F ) . Though NS5 is the most highly conserved flavivirus protein , the high degree of specificity exhibited by UBR4 for DENV NS5 underscores the differences between the various flaviviral NS5 proteins . In 293T cells and primary human dendritic cells , DENV replicates best when UBR4 levels are normal , but when UBR4 levels are reduced , DENV-mediated STAT2 degradation is reduced and DENV replication decreases as a consequence ( Figures 5 and 6 ) . In Vero cells , which do not produce IFN-I [45] , UBR4 depletion does not affect DENV replication unless these cells are treated with exogenous IFN-I ( Figure 5C ) . Furthermore , the DENV replication defect caused by UBR4 knockdown in 293T cells can be decreased by treating the cells with antibodies that block the IFN-I receptor and decrease IFN-I signaling ( Figure 5D ) . The DENV replication defect seen in UBR4-knockdown 293Ts and MDDCs can be explained by an increase in ISG levels in DENV-infected UBR4-knockdown cells versus DENV-infected control cells ( Figure 5E and Figure 6 ) . Thus , in the absence of IFN-I , there is no need for DENV to antagonize IFN-I signaling and cellular levels of UBR4 are irrelevant for DENV replication . However , upon activation of the IFN-I signaling pathway , UBR4 becomes necessary for DENV replication . Reducing STAT2 levels is essential for DENV to preempt the establishment of a cellular antiviral state , thus ensuring its efficient replication . Antagonism of IFN signaling is one of the factors responsible for the limited host tropism of DENV to human and nonhuman primates . DENV does not replicate to high levels or induce disease in IFN-competent mice [44] , [49] . Our previous results indicated that the cellular machinery needed for DENV replication in murine cells is in place but is limited by the inability of NS5 to associate with murine STAT2 and inhibit murine IFN-I signaling [44] . Other blocks such as the type II IFN pathway also diminish DENV replication in mice , but the IFN-I signaling pathway restricts early replication [44] , [50] . Here we show that DENV NS5 associates with murine UBR4 in murine cells . This is in keeping with our previous results [44] , and suggests that the development of a genetically-modified mouse that expresses a functional human STAT2 in place of its murine counterpart should allow increased DENV replication . We predict , therefore , that DENV NS5 will mediate human STAT2 degradation in these mice by co-opting mouse UBR4 . Such a mouse might provide the basis for the development of an immune-competent mouse model of DENV infections . The 600 kDa large UBR4 is highly conserved and found in organisms as diverse as mammals , insects , plants and worms . It belongs to the N-recognin family , which contains proven and predicted E3 ligases that recognize and degrade proteins containing destabilizing N termini . The seven members of the UBR family , UBR1 to UBR7 , encode a 70-amino-acid zinc finger motif known as the UBR box , which is necessary for substrate recognition [42] . The better-characterized members of the UBR family are UBR1 , UBR2 and UBR5 . UBR1 and UBR2 are RING domain-containing N-recognins , which recognize N-end rule substrates and target them for degradation [42] . UBR1 and UBR2 are also involved in N-end-rule-independent quality control protein degradation [51] . UBR5 is a HECT-domain containing E3 ligase that binds N-end rule substrates [42] , but can also target non-N-end rule substrates like E6AP for degradation [52] . UBR4 contains neither a HECT nor a RING domain . A dearth of UBR4 literature exists because of the difficulty that manipulating the UBR4 gene presents . The UBR4 gene contains 106 exons , and produces multiple splice variants that conceivably have different functions . UBR4 forms a chromatin scaffold when bound to retinoblastoma protein ( Rb ) in the nucleus , and it also influences cytoskeleton organization by binding clathrin in the cytoplasm [53] . Both of these are structural roles for which no N-end rule or other E3 ligase activities have been detected . A second virus , human papilloma virus , is known to exploit UBR4's role in cellular morphology to initiate anchorage-independent growth and cellular transformation [54] , [55] . Although UBR4 is part of a family of UBR E3 ligases involved in the N-end rule pathway , the involvement of the N-end rule in the NS5-dependent degradation of STAT2 seems unlikely . Our group has previously demonstrated that the identity of NS5's first residue is not relevant for STAT2 degradation as long as the precursor is correctly processed [13] . In addition , we show that residues T2 and G3 of NS5 are critical for binding to UBR4 and for mediating STAT2 degradation , but they are considered to be stabilizing residues within the N-end rule . This does not exclude UBR4 from having E3 ligase activity that is independent of the N-end rule . Though it lacks an obvious catalytic domain such as the HECT or RING domains , UBR4 contains a cysteine-rich domain ( CRD ) that is unique to the UBR4 group . It is currently unknown if CRD functions as a ligase domain . Our experiments with the N-terminal region of UBR4 suggest that domains from the C terminus , which contain the CRD , are necessary for its function in DENV-mediated STAT2 degradation . Finally , we propose two working models: one based on the hypothesized UBR4 E3 ligase catalytic activity , and another which postulates a scaffolding role for UBR4 based on its described interactions with clathrin and retinoblastoma protein ( Figure 7 ) . Efforts are currently being made to clone and express the predicted UBR4 isoforms so as to further evaluate the function of UBR4 in DENV-mediated STAT2 degradation , and to explore its potential as a target for rationally-designed DENV therapeutics .
293T , Hepa1 . 6 , U6A , BHK and Vero cells were maintained in DMEM ( Life Technologies ) supplemented with 10% fetal bovine serum ( Life Technologies ) and 1% penicillin/streptomycin mix ( Life Technologies ) . C6/36 cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum ( Life Technologies ) . Hepa1 . 6 cells were kindly provided by Matthew Evans ( Mount Sinai School of Medicine , New York , NY ) . U6A cells were a kind gift of George Stark ( Lerner Research Institute , Cleveland , OH ) and were previously described [43] . High-titer stocks of DENV1 , DENV2 , DENV3 , DENV4 , yellow fever virus ( YFV-17D ) and encephalomyocarditis virus ( EMCV ) were obtained by passage in C6/36 cells , BHK cells , and Vero cells , respectively . pCAGGS-CTAP was a kind gift from Luis Martinez-Sobrido ( University of Rochester ) . A gene cloned into pCAGGS-CTAP produces a fusion protein that is C-terminally tagged with a TAP tag: calmodulin binding protein followed by two tobacco etch virus ( TEV ) cleavage sites followed by a protein A tag . The sequences of the primers used for the construction of the RFP-ubiquitin-NS5 fragment that was cloned into pCAGGS-CTAP are available upon request . The primers sequences used for cloning UBR4-NT ( 1–2233 of UBR4 , NCBI Accession Q5T4S7 ) into pCDNA6 are also available upon request . All other viral gene expression constructs were cloned into pCAGGS-HA and were described previously [13] , [15] . The Flag-tagged STAT1 , STAT2 and chimeric STAT constructs were previously described [44] . All cells were transfected using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's protocol . 293T cells were transfected at a ratio 1∶2 ( µg plasmid DNA: µL Lipofectamine 2000 ) while Vero , Hepa1 . 6 and U6A cells were transfected at a ratio 1∶3 . Cells were lysed for tandem affinity purification ( TAP ) or immunoprecipitation two days post transfection . For tandem affinity purification , cells were lysed in TAP buffer ( 25% glycerol , 50 mM Tris HCL pH 8 , 0 . 5% NP40 , 200 mM NaCl , 1 mM β-mercaptoethanol , protease inhibitor cocktail ( Roche ) . Lysates were spun at 15 , 000 g for 10 minutes and the supernatant was incubated with IgG beads ( Roche ) for 4 hours then washed with TAP buffer . The beads were then incubated with TEV buffer ( TAP buffer containing 0 . 5 mM EDTA , 1 mM DTT units ) and 50 units AcTEV enzyme ( Invitrogen ) overnight . The beads were spun at 15 , 000 g for 10 minutes then the supernatant was applied to calmodulin beads ( Roche ) in a calmodulin bead ( CB ) buffer ( TAP buffer containing 4 mM CaCl2 and 2 mM imidazole ) for 8 hours , then washed in CB buffer . The protein was eluted from the calmodulin beads by boiling for five minutes in Laemmli sample buffer ( BioRad ) . For immunoprecipitation , cells were lysed in TAP buffer then incubated for two hours with anti-FLAG or anti-HA beads ( #F2426 and #E6779 respectively , Sigma-Aldrich ) or for 4 hours with rabbit anti-STAT2 antibody ( Santa Cruz ) or mouse anti-GFP antibody ( Abcam ) followed by 2 hours of protein A-agarose ( Roche ) or protein G-agarose ( Roche ) , respectively . The beads were washed with TAP buffer then the protein was eluted from the beads by boiling for five minutes in Laemmli sample buffer ( BioRad ) . Proteins lysates were boiled with Laemmli sample buffer and resolved on 4–15% or 7 . 5% gels ( BioRad ) and then transferred to PVDF membrane ( Millipore ) by standard methods . Membranes were blocked with 3% BSA in TBS-Tween ( 20 mM Tris-HCl , pH 7 . 4; 150 mM NaCl; 1% Tween ) and then incubated with antibodies and subjected to western blot . Benchmark Protein Ladder ( Invitrogen ) was used to depict the size of protein bands . The primary antibodies used in this study were: rabbit anti-human STAT2 ( sc-476 , Santa Cruz ) , rabbit anti-mouse STAT2 ( 4597 , Cell Signaling ) , rabbit anti-STAT1 ( 610120 , BD Biosciences ) , mouse anti-β-tubulin ( T0198 , Sigma-Aldrich ) , mouse anti-HA ( H9658 , Sigma-Aldrich ) , mouse anti-Flag ( F3165 , Sigma-Aldrich ) , mouse anti-V5 ( R960-25 , Invitrogen ) , rabbit anti-UBR4 ( ab86738 , Abcam ) , HRP-linked anti-GAPDH ( ab9385 , Abcam ) , rabbit anti-UBR5 ( ab70311 , Abcam ) , and rabbit anti-NS5 [13] . The secondary antibodies used in this study were HRP-linked anti-mouse IgG ( #NA931V , GE Healthcare ) and HRP-linked anti-rabbit IgG ( #NA934V , GE Healthcare ) . Where indicated , quantification of western blots was done by using Image J to compare the ratio of UBR4 ( seen as two bands or one band based on the resolution of the tris-glycine gel used ) to NS5 . To analyze the intracellular localization of endogenous UBR4 and DENV NS5 , Vero cells that had been grown on glass cover slips were transfected with 1 µg of the indicated plasmids . After 24 hours post infection , cells were fixed and permeabilized for 30 minutes with ice cold methanol acetone ( 1∶1 , v/v ) and 0 . 5% NP-40 , then washed with PBS . Following PBS washes , cells were blocked in blocking buffer ( 0 . 2% cold waterfish gelatin ( Sigma-Aldrich , USA ) and 0 . 5% BSA in PBS ) for 1 hour at room temperature ( RT ) , and stained with primary antibodies ( anti-UBR4 at a 1∶100 dilution , and anti-HA at a 1∶1000 dilution ) overnight at 4°C . The cells were washed in PBS and incubated with secondary antibodies to Alexa Fluor 488 and Alexa Fluor 555 ( Invitrogen , USA ) at 1∶500 dilution in blocking buffer for 1 hour at RT . Nuclear chromatin staining was performed by incubation in blocking solution containing 0 . 5 mg/ml 4′ , 6-diamidino-2-phenylindole , DAPI ( Sigma-Aldrich ) . Cells were washed and coverslips mounted using Prolong antifade reagent ( Invitrogen ) . Images were captured using a Leica SP5-DM confocal microscope at the Microscopy Shared Research Facility at Mount Sinai School of Medicine . The 293T and Vero cell lines stably expressing non-silencing shRNA or shRNA against UBR4 were made by infecting cells with shRNA-encoding lentiviruses ( according to the manufacturer's protocol ) and selecting cells with puromycin ( 1 µg/ml for 293T cells and 5 µg/ml for Vero cells ) for two weeks before DENV , YFV or EMCV infection . The lentiviruses used to make 293T shUBR4 clones were purchased from Open Biosystems . Lentivirus 1 ( Clone ID: V3LHS_318553; target sequence: CGCTTCGACTTCATGCTCT ) targets nucleotides 11132–11150 of UBR4 . Lentivirus 2 ( Catalog #: V3LHS_318554; target sequence: CGGATCAGCTCCTATGTCA ) targets nucleotides 3140–3158 of UBR4 . Lentivirus 3 ( Catalog #: V3LHS_318555; target sequence: AGGTTTTTGTCTACAATGA ) targets nucleotides 2357–2375 of UBR4 . The non-silencing control lentivirus was catalog number RHS4348 . The Vero shUBR4 clone was made using lentivirus 1 , while the Vero shControl was made using non-silencing control lentivirus . Cells were infected at the indicated multiplicity of infection ( MOI ) and maintained in DMEM with 10% FBS . For exogenous IFN-I treatment of Vero cells , 1000 units/ml IFNβ ( PBL Interferon Source ) were added at 6 hours post infection . For western blotting , cells were lysed with TAP buffer at each time point and the lysates were clarified by centrifugation then boiled in Laemmli sample buffer . For virus titration , cells and media were frozen at each time point and clarified by centrifugation . DENV and YFV titers were measured by plaque assay on BHK-21 cells , and EMCV titers were measured by plaque assay on Vero cells . Peripheral blood mononuclear cells were isolated from buffy coats of healthy human donors by Ficoll density gradient centrifugation ( Histopaque , Sigma Aldrich ) as previously described [33] . Buffy coats were obtained from the Mount Sinai Blood Donor Center and New York Blood Center . Briefly , CD14+ cells were purified using anti-human CD14 antibody-labeled magnetic beads and iron-based MiniMACS LS columns ( Miltenyi Biotech ) . After elution from the columns , 2×105 cells were plated in 96-well plates and transduced with VSV-G pseudo-typed SIV VLPs ( pSIV3+ , an SIV gag-pol expression plasmid containing Vpx ) and lentiviral control or UBR4-specific shRNA vectors for 3 hours by spinoculation in the presence of 2 µg/mL polybrene ( Sigma ) with sufficient viruses to transduce >95% of the cells . Subsequently , cells were washed , resuspended in DC medium ( RPMI medium [Invitrogen] , 10% fetal calf serum [HyClone] , 100 U/ml penicillin , and 100 µg/ml streptomycin [Invitrogen] ) supplemented with 500 U/ml human granulocyte-macrophage colony-stimulating ( Peprotech ) , and 1 , 000 U/ml human interleukin-4 ( hIL-4; Peprotech ) , and incubated for 5 days at 37°C . At 5 days post transduction , MDDCs were either mock infected or infected with DENV2 at an MOI of 3 . At 12 hours post infection ( hpi ) cells were harvested for qPCR analysis , and at 48 hpi supernatants were collected for titration of virus levels by plaque assay on BHK-21 cells . Cells were also harvested for cytometry analysis . Total RNA was isolated from samples using the RNeasy kit ( Qiagen ) and subjected to DNase digestion with Turbo DNase ( Ambion ) . Reverse transcription was performed using the high capacity cDNA reverse transcription kit ( Applied Biosystems ) . qPCR was performed in 384-well plates in triplicates using SYBR green I master mix ( Roche ) in a Roche LightCycler 480 . Relative mRNA values were calculated using the ΔΔCt method using 18S rRNA as internal control and plotted as fold change by normalizing to mock-control samples . UBR4 qPCR primers: Forward = GGTGTTCCAGAGGCTAGTGATC; Reverse = CCAACTGCTTCTGCGGTTCCTT ISG15 qPCR primers: Forward = TCCTGGTGAGGAATAACAAGGG; Reverse = GTCAGCCAGAACAGGTCGTC RIG-I qPCR primers: Forward = GGCATGTTACACAGCTGACG; Reverse = TGCAATATCCTCCACCACAA ISG54 qPCR primers: Forward = ATGTGCAACCTACTGGCCTAT; Reverse = TGAGAGTCGGCCCATGTGATA 18S RNA qPCR primers: Forward = GTAACCCGTTGAACCCCATT; Reverse = CCATCCAATCGGTAGTAGCG DENV-infected DCs were fixed and permeabilized with Cytofix and Cytoperm reagent ( BD Pharmingen ) according to the manufacturer's recommendations . Then , cells were stained with 4G2 ( ATCC ) , a mouse monoclonal antibody specific for the E protein , as a primary antibody and a FITC-labeled anti-mouse antibody as a secondary antibody . Flow cytometry was performed using a FACScan flow cytometer ( Becton Dickinson ) and analyzed with FlowJo software . | Dengue virus ( DENV ) is the leading cause of mosquito-borne viral illness and death in humans . At present , there are no vaccines and no specific antiviral therapeutics to prevent or treat DENV infections . We previously described that the NS5 protein of DENV inhibits type I interferon signaling in virus-infected cells by mediating STAT2 degradation . This property allows DENV to overcome the antiviral effects of type I interferon , contributing to viral replication in the host . We have now obtained new insight into the mechanism by which DENV NS5 induces STAT2 degradation . NS5 bridges STAT2 with the cellular protein UBR4 , a member of a family of predicted E3 ligases , resulting in UBR4-mediated STAT2 degradation . Elimination of UBR4 or mutations in NS5 that prevent its binding to UBR4 prevent NS5 from inducing STAT2 degradation . Importantly , UBR4 is required for optimal DENV replication in the presence of a competent type I interferon system . Our data demonstrate the requirement of a host factor , UBR4 , for DENV to overcome the antiviral interferon response . This information might be important for the design of specific DENV inhibitors that prevent dengue virus from evading innate immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"global",
"health",
"immunology",
"biology",
"microbiology",
"public",
"health"
] | 2013 | Dengue Virus Co-opts UBR4 to Degrade STAT2 and Antagonize Type I Interferon Signaling |
In the hierarchy of cellular targets damaged by ionizing radiation ( IR ) , classical models of radiation toxicity place DNA at the top . Yet , many prokaryotes are killed by doses of IR that cause little DNA damage . Here we have probed the nature of Mn-facilitated IR resistance in Deinococcus radiodurans , which together with other extremely IR-resistant bacteria have high intracellular Mn/Fe concentration ratios compared to IR-sensitive bacteria . For in vitro and in vivo irradiation , we demonstrate a mechanistic link between Mn ( II ) ions and protection of proteins from oxidative modifications that introduce carbonyl groups . Conditions that inhibited Mn accumulation or Mn redox cycling rendered D . radiodurans radiation sensitive and highly susceptible to protein oxidation . X-ray fluorescence microprobe analysis showed that Mn is globally distributed in D . radiodurans , but Fe is sequestered in a region between dividing cells . For a group of phylogenetically diverse IR-resistant and IR-sensitive wild-type bacteria , our findings support the idea that the degree of resistance is determined by the level of oxidative protein damage caused during irradiation . We present the case that protein , rather than DNA , is the principal target of the biological action of IR in sensitive bacteria , and extreme resistance in Mn-accumulating bacteria is based on protein protection .
The amount of DNA damage caused by a given dose of γ-radiation for resistant and sensitive bacteria is very similar [1 , 2] . Yet , the range of ionizing radiation ( IR ) resistances is large [1–3] , with a factor of 200 separating the most-resistant from the most-sensitive species [1] . For example , Deinococcus radiodurans can survive levels of IR ( 10 kGy ) that induce approximately 100 DNA double-strand breaks ( DSBs ) per genome , whereas Shewanella oneidensis is killed by levels of IR ( 0 . 07 kGy ) that result in less than 1 DSB per genome [1] . We have reported a relationship between intracellular Mn/Fe concentration ratios and bacterial survival following exposure to IR , in which the most-resistant cells contained about 300 times more Mn and about three times less Fe than the most-sensitive cells [1] . Furthermore , restricting Mn ( II ) during growth of D . radiodurans significantly lowered the Mn content of wild-type cells , and IR resistance to levels quantitatively similar to several highly sensitive D . radiodurans DNA repair mutants [1] . However , the nature of Mn-facilitated IR resistance was undefined , and the question of why many bacteria that encode a complement of repair functions are killed by doses of IR that cause little DNA damage has not been resolved [1 , 4 , 5] . Broad-based bioinformatic and experimental studies have converged on the conclusion that D . radiodurans uses a relatively conventional set of DNA repair and protection functions , but with far greater efficiency than IR-sensitive bacteria [1–12] . Despite these efforts , however , the molecular mechanisms underlying the extraordinary IR resistance of D . radiodurans and other Mn-accumulating bacteria remain poorly understood [3 , 4] . For example , recent work by Zahradka et al ( 2006 ) [7] showed that DNA polymerase I ( PolA ) of D . radiodurans supports very efficient DNA replication at the earliest stages of recovery , and could account for the high fidelity of RecA-dependent DSB fragment assembly [7] . However , IR-sensitive D . radiodurans polA mutants are fully complemented by expression of the polA gene from the IR-sensitive Escherichia coli [12] . The reason why repair proteins , either native or cloned , in D . radiodurans cells function so much better after irradiation than in other organisms is unknown . We show that the amount of protein damage caused by a given dose of γ-radiation for intrinsically resistant and sensitive bacteria is very different . High levels of protein protection during irradiation correlated with high intracellular Mn/Fe concentration ratios and high levels of resistance , whereas proteins in radiation-sensitive cells were highly susceptible to IR-induced oxidation .
In comparison to Fe ( II ) , Mn ( II ) does not significantly react with dioxygen ( O2 ) or hydrogen peroxide ( H2O2 ) at physiological pH values in water . However , Mn ( II ) has been reported to react strongly with superoxide radicals ( O2•− ) . For example , as free ions or when complexed with lactate or succinate , Mn ( II ) can act as a potent scavenger of O2•− , with Mn cycling between the divalent and trivalent states , releasing H2O2 as an intermediate [13] . In contrast , when complexed with bicarbonate , Mn ( II ) catalyzes the disproportionation of H2O2 [14] . Thus , the presence of Mn might affect the relative abundance of reactive oxygen species ( ROS ) generated during irradiation . During the γ-radiolysis of water , solvated electrons ( e−aq ) react rapidly with O2 to form O2•− [15] , which could react with Mn ( II ) and protons ( H+ ) to form H2O2 and Mn ( III ) [13] . Mn redox cycling would occur if Mn ( III ) generated during irradiation was reduced back to Mn ( II ) by an electron donor such as H2O2 , a major and relatively stable product of the radiolysis of water ( Figure 1 ) . The primary oxygen radicals generated in the radiolysis of water are hydroxyl radicals ( HO• ) and peroxyl radicals ( R-O2• ) [15] . On the basis of the different reactivities of HO• and O2•− with DNA and proteins , we tested the ability of Mn ( II ) to scavenge these ROS during irradiation . DNA is readily damaged by HO• , but insensitive to O2•− [16] . In contrast , O2•− damages [4Fe-4S] cluster-containing enzymes such as aconitase [16] , and Mn ( II ) has been reported to protect enzymes from Fe-catalyzed inactivation by O2 in the presence of electron donor systems [17 , 18] . For example , the restriction enzyme BamHI is readily deactivated by ROS generated under aerobic conditions in the presence of Fe ( II ) and ascorbate , but not when O2 is absent or 4 mM Mn ( II ) is present [19] . Consistent with the propensity of HO• , but not O2•− , to damage DNA dissolved in double-distilled de-ionized water ( dH2O ) [15 , 16] , 1% dimethylsulphoxide ( DMSO , a HO• scavenger ) [15] conferred substantial protection on supercoiled plasmid DNA in vitro during aerobic irradiation , but 5 mM Mn ( II ) did not ( Figure 2A ) . We also tested BamHI for its susceptibility to IR damage in vitro ( Figure 2B and 2C ) . The highest IR dose that BamHI could survive and then function after irradiation under aerobic conditions in dH2O was approximately 50 Gy; in 1% DMSO , it was approximately 150 Gy; and in 5 mM MnCl2 , it was approximately 1 , 000 Gy . Since the deactivating IR dose for BamHI that has been irradiated anaerobically in dH2O in the absence of DMSO or 5 mM MnCl2 was approximately 1 , 000 Gy ( Figure 2B ) , we examined whether free Mn ( II ) ions might protect BamHI from O2-dependent modifications during irradiation . In vitro , 5 mM MnCl2 limited IR-induced oxidative protein damage under aerobic conditions in the presence or absence of Fe , as measured by assaying for carbonyl group ( aldehydes and ketones ) generation into proteins at Lys , Arg , Pro , and Thr residues by oxidative reactions [20] ( Figure 2C ) . The level of carbonyl groups in proteins is widely used as a marker of oxidative protein damage and has attracted a great deal of attention due to its irreversible and unrepairable nature [21] . Since not all oxidative modifications lead to carbonyl derivatives , however , the levels of oxidation detected represent minimal values . These results support our model that Mn ( II ) ions might protect proteins by scavenging peroxyl radicals ( O2•− , HO2• , and R-O2• ) and/or H2O2 , but do not scavenge HO• generated during irradiation ( Figure 1 ) . To demonstrate a mechanistic link between solution-phase radiochemistry of Mn ions ( Figure 2 ) and their physiological targets in vivo , we examined IR-induced protein damage in IR-sensitive and IR-resistant bacteria ( Figure 3 ) . Cellular proteins in the most-sensitive bacteria were substantially more vulnerable to IR-induced oxidation than proteins in the most-resistant bacteria ( Figure 3 ) ; and from the pattern of oxidized bands , we infer that not all proteins in sensitive bacteria are equally susceptible to carbonylation . At 4 kGy , high levels of protein oxidation occurred in cells with the lowest intracellular Mn/Fe concentration ratios , whereas no protein oxidation was detected in cells with the highest Mn/Fe ratios ( see bottom of Figure 3 for bacterial IR survival values and Mn/Fe concentration ratios ) . In vitro , proteins from resistant bacteria were readily carbonylated when exposed to IR in the presence of Fe ( Figure 4A ) , which confirmed that proteins in resistant bacteria are not inherently resistant to oxidation . Furthermore , we previously reported that D . radiodurans cells grown in defined rich medium without Mn supplementation ( No-Mn DRM ) were depleted in Mn and , at 10 kGy , displayed a 1 , 000-fold reduction in survival compared to cells with normal Mn concentrations [1] . D . radiodurans cells grown in DRM without Mn were highly susceptible to protein oxidation during irradiation ( Figure 4B ) . In comparison , D . radiodurans cells with normal intracellular Mn concentrations were sensitized to IR and protein oxidation when irradiated at pH 10 . 5 ( Figure 4C ) . Pseudomonas putida cells irradiated anaerobically were equally sensitive to IR and as susceptible to IR-induced protein carbonylation as cells irradiated aerobically ( Figure 4D ) . Thus , high levels of IR-induced protein oxidation in bacteria correlated with IR sensitivity in the presence or absence of atmospheric O2 , and the IR resistance and level of protein oxidation in D . radiodurans cells with normal intracellular Mn concentrations could be controlled exogenously . In vitro , the stoichiometry of intermediates and end products of Mn redox cycling is dependent on the concentration of reactants [13] . For example , Mn ( III ) accumulates if H2O2 becomes limiting , whereas an excess of O2•− and H+ favors the formation of H2O2 [13] ( Figure 1 ) . Using assays based on Rhodazine D , a sensitive colorimetric reagent for measuring O2 and H2O2 , we tested whether or not MnCl2 solutions exposed anaerobically to 10 kGy generated these species ( Figures 5A and S1 ) . Dissolved O2 and H2O2 react with the pale yellow leuco form of Rhodazine D to produce a rose color , with the color proportional to the dissolved O2 or H2O2 concentration . Color development in an Ar-purged solution tested after irradiation ( Figure 5A , column III ) , but not upon re-purging with Ar ( Figure 5A , column V ) , indicated O2 formation . Color development in an Ar-purged solution tested after irradiation and upon re-purging with Ar ( Figure 5A , column V ) , but not following catalase treatment ( Figure 5A , column VI ) , indicated H2O2 accumulation . Consistent with the existence of a threshold concentration of Mn ( II ) needed for Mn redox cycling [13] under in vitro irradiation were ( 1 ) exposure of anaerobic dH2O or MnCl2 solutions at concentrations below 0 . 1 mM to 10 kGy did not generate detectable levels of O2 or H2O2 ( Figure 5A and 5B ) ; ( 2 ) exposure of anaerobic MnCl2 solutions at intermediate concentrations ( 0 . 1–10 mM ) to 10 kGy generated H2O2 ( Figure 5A ) ; and ( 3 ) from initially anaerobic conditions , exposure of MnCl2 solutions at high concentrations ( 1 M ) to 10 kGy yielded O2 , but not H2O2 ( Figure 5A ) , and copious Mn dioxide precipitates ( Figure S2 ) ; we infer that the accumulation of Mn dioxides was caused by a shortage of H2O2 [13] , which is decomposed by γ-radiolysis [15] . The obligate aerobic D . radiodurans accumulates 2 mM or greater Mn ( 1 × 105 Mn atoms/cell , assuming an average cell volume of 6 . 5 × 10−2 μm3 ) [1] , which is present in cells as Mn ( II ) ( Figure 6A ) [1] , and the facultative anaerobic , radioresistant bacterium Lactobacillus plantarum [1 , 22] accumulates 20–25 mM Mn ( II ) [23] . For an irradiated cell containing 2–25 mM Mn engaged in catalytic Mn redox cycling [13] , intracellular O2•− , which does not easily cross biological membranes [16 , 24] , might be reduced to H2O2 , which is membrane permeable and could diffuse out of the cell and accumulate ( Figure 1 ) . When exposed to 10 kGy as cell suspensions in dH2O at 0 °C and pre-conditioned to be anaerobic , H2O2 was released by D . radiodurans ( ∼2 × 10−5 M ) and L . plantarum ( ∼6 × 10−5 M ) , in which the presence of H2O2 after irradiation was confirmed by catalase-treatment ( Figure 5A ) ; notably , H2O2 was consumed by the cells upon incubation at 32 °C ( Figure 5A ) . We infer that H2O2 was produced by intracellular Mn ( II , III ) redox cycling . In contrast , detectable levels of H2O2 or O2 were not released by non-irradiated D . radiodurans or L . plantarum control samples , nor by irradiated S . oneidensis ( Figure 5A ) , which accumulates substantially more Fe than Mn and is extremely sensitive to IR [1 , 5] . Reducing the equilibrium concentration of H+ with hydroxide ions ( OH− ) is predicted to limit Mn redox cycling [13] and the accumulation of H2O2 ( Figure 1 ) . In testing the possibility that survival of irradiated D . radiodurans is vulnerable to conditions that limit intracellular acidification , we first established that IR-driven Mn redox cycling could be inhibited in vitro by increasing the pH . In vitro , exposure to IR of anaerobic 10 mM MnCl2 at pH 9 or higher yielded O2 but not H2O2 , whereas H2O2 was formed at pH values below 9 ( Figure 5B ) . Consistent with the prediction that Mn redox cycling can be inhibited in vivo by OH− , D . radiodurans irradiated at pH 10 . 5 did not release H2O2 or O2 ( Figure 5A ) . In vivo , the 10% survival value ( D10 ) [1] of D . radiodurans cells grown , irradiated , and recovered at pH 7 was 16 kGy , whereas the D10 was 6 kGy for cells grown at pH 7 , irradiated at pH 10 . 5 , and recovered at pH 7 ( Figure 6B , top ) ; survival of non-irradiated D . radiodurans control samples held at a pH of 11 or lower for 16 h was approximately 100% ( Figure 6B , bottom ) . In vivo , the pH-dependent loss in IR resistance of D . radiodurans ( Figure 6B , top ) correlated with a substantial increase in oxidative protein damage during irradiation ( Figure 4C ) . Thus , at pH values at which IR-driven Mn redox cycling was inhibited in vitro ( Figure 5B ) , the IR resistance of D . radiodurans was significantly decreased , and the cells were highly susceptible to IR-induced protein oxidation ( Figure 4C ) . Because the formation of ROS during irradiation is extremely rapid [15] , an intracellular protection system which is ubiquitous , but not highly dependent on the induction of enzymes , stage of growth , or temperature over a range at which cells are metabolically active , could provide a selective advantage to the host in some environments . In this context , we examined the intracellular distribution of Mn and Fe in D . radiodurans cells using X-ray fluorescence ( XRF ) microprobe analysis [25] ( Figures 6C and S3 ) . Within a representative diplococcus , whereas Mn ( II ) was globally distributed , Fe was partitioned largely outside the cytoplasm in a region overlapping the septum between dividing cells ( Figure 6D ) .
We previously demonstrated a critical role for the accumulation of Mn ( II ) in D . radiodurans in a mechanism toward surviving IR that is not dependent on Mn-SOD ( Mn-dependent superoxide dismutase ) [1] . D . radiodurans contains four to ten identical copies of its genome per cell , and when irradiated to a dose of 10 kGy , generates more than 400 genomic DSB fragments per cell [1 , 8 , 26] . Yet , this amount of DNA damage in D . radiodurans does not typically lead to cell death [1 , 8] . Bioinformatic and experimental reports generally support that genome configuration and copy number , and enzymatic protection and repair functions of D . radiodurans do not have unique properties that are essential or prerequisite for expression of the extreme-resistance phenotype [1–12] . For example , D . radiodurans DNA repair and protection genes do not differ greatly from their counterparts in the IR-sensitive S . oneidensis , P . putida , or E . coli [1 , 6]; several E . coli DNA repair genes have been shown to fully restore corresponding radiation-sensitive D . radiodurans mutants to wild-type levels of D . radiodurans resistance [12 , 27]; Mn-SOD is not needed for survival of D . radiodurans following acute irradiation or growth under 50 Gy/h [1 , 11]; non-homologous end joining of D . radiodurans chromosomal DSB fragments following IR is not observed [28]; and the products of interchromosomal recombination in D . radiodurans following irradiation are consistent with the canonical version of the DSB repair model [8] . Over the last decade , several hypotheses to reconcile these findings have built on the idea that D . radiodurans might use mechanisms that restrict the diffusion of DNA DSB fragments produced following irradiation , to facilitate repair [3 , 29] . However , transmission electron microscopy [TEM] of Mn-depleted , radiosensitive D . radiodurans displayed normal levels of chromosomal condensation [1] , and cryoelectron microscopy of vitreous sections of D . radiodurans supports the conclusion that DNA fragments in D . radiodurans are mobile and that the arrangement of its nucleoids does not play a key role in radioresistance [10 , 30] . Consistently , a series of earlier molecular studies on irradiated D . radiodurans cells showed high levels of recombination between homologous DSB fragments originating from widely separated genomic locations [8 , 28 , 31] . Evidence presented here supports the idea that the extreme-resistance phenotype of D . radiodurans and other bacteria with high intracellular Mn/Fe concentration ratios is dependent on a mechanism that protects proteins from oxidative damage during irradiation , which could offset the need for highly specialized cellular repair systems . Previous work has shown that the linear density of DSBs introduced into genomic DNA by a given dose of IR in extremely resistant and sensitive bacteria is essentially the same [1 , 2] . In contrast , we find that protein damage is quantifiably related to bacterial radioresistance ( Figures 3 , 4B , and 4C ) . The most-sensitive cells had very low Mn/Fe concentration ratios and were highly susceptible to IR-induced protein oxidation , whereas the most-resistant cells had high intracellular Mn/Fe ratios and were relatively insusceptible to protein oxidation . Although the mechanism by which Mn protects proteins during irradiation remains unknown , our results provide insight into how Mn redox cycling could attenuate the detrimental effects of Fe redox cycling . Scavenging of ROS in IR-resistant bacteria may be linked to both the presence of Mn and relatively low cytosolic levels of Fe ( Figure 6C and 6D ) [1] . Our observation that resistant bacteria released H2O2 during irradiation , but sensitive bacteria did not ( Figure 5A ) , is consistent with the idea that Fe redox cycling is limited in Mn-accumulating cells . Most bacteria accumulate near-millimolar concentrations of intracellular Fe , primarily for assembly of Fe-S and haem proteins [1 , 32 , 33] . However , resistant bacteria typically encode fewer proteins with Fe-S domains than sensitive bacteria [4] . Since IR-induced ROS likely damage exposed Fe-S clusters , releasing Fe ( II ) [16 , 32] , Fe-laden sensitive cells might be predisposed to Fe redox cycling reactions during irradiation . Since 5 mM Mn ( II ) does not significantly scavenge HO• ( Figure 2A ) , Mn redox cycling likely does not protect cells from HO• generated either directly by water radiolysis or indirectly by the Fenton ( Fe ( II ) ) reaction ( Figure 1 ) [15] . However , the Haber-Weiss ( Fe ( III ) ) reaction ( Figure 1 ) [15] generates O2 , which under IR would give rise to O2•− and other peroxyl radicals ( R-O2• ) ( Figure 1 ) [15] . Compared to HO• , O2•− is relatively unreactive with a large number of compounds including DNA , but can undergo chain reactions leading to organic hydroperoxides [15] , which decompose in the presence of Fe to give new radicals , including oxidizing alkoxyl radicals that are more reactive than peroxyl radicals [15 , 34] . Although intracellular Mn ( II ) does not protect cells from DNA damage caused by HO• [1] , scavenging of simple peroxyl radicals by Mn redox cycling might prevent the propagation of secondary reactions that ultimately damage proteins [15 , 16 , 34] . The exceptionally high catalase activities of D . radiodurans [11] might be expected to favor the accumulation of Mn ( III ) ( Figure 1 ) , a strong oxidant capable of damaging cell components [13] . We did not detect significant levels of Mn ( III ) or Mn ( IV ) in irradiated or non-irradiated D . radiodurans by X-ray-absorption near-edge structure ( XANES ) spectroscopy ( Figure 6A ) , indicating that Mn ( III ) might also be reduced by other mechanisms in vivo . Notably , when Mn is complexed with succinate or lactate in vitro , the efficiency of Mn redox cycling is greatly increased , such that Mn ( III ) reactivity is similar to that of Mn ( II ) complexes and re-reduction of Mn ( III ) by O2•− might occur [13] . In this context , metabolic pathway switching in D . radiodurans cells immediately after irradiation has been reported to favor the production of succinate via up-regulation of the glyoxylate bypass of its tricarboxylic acid ( TCA ) cycle , and down-regulation of degradative steps of the TCA cycle [4 , 5 , 9] . In bicarbonate/CO2 buffer , Mn ( II ) is reported to catalyze the oxidation of free amino acids such as leucine and alanine by H2O2 and the dismutation of H2O2 [35] . Thus , complexes containing Mn and amino acids or organic acids might scavenge H2O2 in addition to O2•− in cells exposed to IR . An interesting feature of the systems for energy production in D . radiodurans is that , unlike most other free-living bacteria , it uses the vacuolar type of proton ATP synthase instead of the F1F0 type [6] . Vacuolar ( V ) -type H+-ATPases are typical of eukaryotes and archaea , and central players in intracellular acidification [36] , which might facilitate Mn redox cycling by providing H+ ( Figure 1 ) . In this context , our findings generally support bioinformatic studies by Karlin and Mrazek in 2001 [37] , who proffered a new explanation for the resistance of D . radiodurans contingent on a role of predicted highly expressed ( PHX ) genes for proteases , the glyoxylate bypass of the TCA cycle , ABC-type transporters of amino acids and Mn , and ( V ) -type ATPases . The speciation , distribution , and relatively high concentration of Mn in D . radiodurans [1] ( Figure 6A and 6C ) support the idea that Mn ( II ) could provide immediate cytosolic protection from O2•− , and facilitate removal of H2O2 from cells exposed to IR . Further , electron-dense granules ( EDGs ) ( Figure 6C , TEM , circular dark ∼0 . 2 μM inclusions ) , which are frequently observed in electron microscopy images at the center of D . radiodurans nucleoids [1 , 4 , 10] , were associated with the highest regional Mn concentrations ( 200 parts per million [ppm]; 3 . 6 mM ) ( Figure 6D ) , perhaps to protect enzymic DNA repair functions , many of which are dependent on redox-active [4Fe-4S] clusters [38] . Importantly , our findings do not preclude the possibility that intracellular Mn ( II ) also prevents lipid peroxidation in cell membranes . In this context , however , the lowest regional Mn concentrations ( 50 ppm ) were associated with the cell envelope [6] ( Figures 6C and S3 ) , indicating that Mn ( II ) predominantly protects the cytosol; and earlier reports strongly support the idea that lipid peroxidation can be dissociated from lethal damage in irradiated mammalian and irradiated bacterial cells [34 , 39] . During recovery of irradiated D . radiodurans , additional damage might be limited by secondary antioxidant defenses , including attendant cellular responses that limit the production of metabolism-induced ROS [4 , 9] , and degradation of oxidized proteins by the expanded family of subtilisin-like proteases [6] . The Mn content of bacteria [1] might also determine the amount of protein damage caused in cells exposed to other oxidative stress-inducing conditions , including desiccation [1 , 6 , 24] and ultraviolet ( UV ) radiation [5 , 12] , and xenobiotic agents such as Cr ( VI ) and mitomycin-C ( MMC ) that elicit redox-related toxicity [40] . Chromosomal DNA is an indispensable molecule whose integrity must be conserved following exposure of a cell to IR to ensure survival [15] , such that the functionality of DNA repair and replication systems ultimately determines if an irradiated cell lives or dies , even for the most IR-resistant bacteria [6 , 12 , 27] . Our findings that IR-induced cellular protein damage ( Figures 3 and 4 ) , but not DNA damage [1 , 2] , is quantifiably related to radioresistance , and intracellular Mn/Fe concentration ratios could help explain why bacteria that encode a similar repertoire of DNA repair functions display such large differences in IR resistance [1 , 3 , 5] . Specifically , we propose that redox cycling of Mn ( II ) that is accumulated in resistant bacteria [1] protects proteins from oxidation during irradiation ( Figure 1 ) , with the result that enzyme systems involved in recovery survive and function with great efficiency . This could explain why the polA gene of E . coli fully complements IR- , UV- and MMC-sensitive D . radiodurans polA mutants [12] . In comparison , we attribute the high level of radiation sensitivity of Fe-rich , Mn-poor bacteria to their susceptibility to global Fe-mediated oxidative protein damage during irradiation under aerobic or anaerobic conditions ( Figure 4D ) . Oxidative modification of proteins by IR could disrupt cellular functions involved in DNA repair either by loss of catalytic and structural integrity or by interruption of regulatory pathways , which in extremely radiation-sensitive cells might render protein damage lethal before significant DNA damage has accumulated [5] . At sublethal IR doses in sensitive cells , oxidatively damaged DNA repair enzymes would be expected to passively promote mutations by misrepair . Oxidized proteins , however , might also actively promote mutation by transmitting damage to other cellular constituents , including DNA [34 , 41] . In conclusion , our data provide a novel framework for understanding how intracellular Mn and Fe contribute to IR resistance , which is important since these findings may come to affect models of radiation toxicity , as well as approaches to control recovery from radiation injury [42] , including the development of systems for delivery into cells of Mn-based radioprotective complexes or Fe-based radiosensitizers [43] .
The wild-type strains used were as follows: D . radiodurans ( ATCC BAA-816 ) ; D . geothermalis ( DSM 11300 ) ; S . oneidensis ( MR-1 ) ( ATCC 700550 ) ; P . putida ( ATCC 47054 ) ; Enterococcus faecium ( ATCC 19434 ) ; L . plantarum ( ATCC 14917 ) ; and E . coli ( K-12 ) ( MG1655 ) . Strains were cultured aerobically in undefined liquid rich medium [1] ( TGY: 1% Bacto-tryptone , 0 . 1% glucose , 0 . 5% yeast extract ) ( pH 7 ) at 32 °C to an optical density at 600 nm ( OD600 ) of 0 . 9–1 . 0 , unless indicated otherwise . For anaerobic irradiations , 6-ml 0–1 , 000 mM MnCl2 solutions in dH2O ( pH ∼6 ) , or approximately 1 × 1010 bacterial cells ( ∼5 × 109 diplococci ) re-suspended in 6-ml dH2O , were transferred to Quick-Seal ultracentrifuge tubes ( 13 cm3 ) ( Beckman Instruments , Palo Alto , California , United States ) , unless indicated otherwise . The tubes were purged at room temperature with ultra-high purity Ar ( 99 . 999% ) ( ValleyNational Gases , Frederick , Maryland , United States ) for 5 min ( 200 cm3/min ) , sealed anaerobically , and irradiated on ice ( 0 °C ) at 1 . 8 Gy/s ( 60Co , Model 109; J . L . Shepard and Associates , San Fernando , California , United States ) . The pUC19 assay was performed as follows . Supercoiled pUC19 ( 1 mg/ml ) ( New England Biolabs , Ipswich , Massachusetts , United States ) was diluted ( 1/25 ) in dH2O , 1% DMSO , or 5 mM MnCl2; 50-μl aliquots of each of the three pUC19 dilutions were irradiated ( 60Co ) at 0 °C to the indicated doses . A total of 88 ng of each IR-treated pUC19 sample was subjected to agarose ( 0 . 9% ) gel electrophoresis in 1 × TBE ( Tris , borate , EDTA buffer ) and 250-ng/ml ethidium bromide , at 47 V for 14 h . Linearized plasmid ( Lp ) ( pUC19 + BamHI ) was used as a marker in gels containing IR-treated supercoiled pUC19 . The BamHI assay was performed as follows . BamHI ( 700 , 000 U/ml ) ( New England Biolabs ) was diluted ( 1/1 , 500 ) in dH2O , 1% DMSO , or 5 mM MnCl2; 1 , 000-μl aliquots of each of the three BamHI dilutions were irradiated ( 60Co ) aerobically at 0 °C to the indicated doses . For anaerobic BamHI irradiations , 5-ml aliquots of diluted BamHI/dH2O were transferred to separate Quick-Seal ultracentrifuge tubes ( 13 cm3 ) ( Beckman Instruments ) , purged at room temperature with ultra-high purity Ar ( 99 . 999% ) ( Valley National Gases ) for 5 min ( 200 cm3/min ) , and sealed anaerobically before irradiation . A total of 40 μl ( 23 units ) of each IR-treated BamHI sample was transferred to separate reaction mixes ( final volume , 60 μl ) containing 250-ng λ-phage DNA ( New England Biolabs ) , 50 mM NaCl , 10 mM Tris-HCl ( pH 7 . 9 ) , 10 mM MgCl2 , and 1 mM dithiothreitol . BamHI/λ DNA reactions were incubated for 1 h at 37 °C , followed by agarose ( 0 . 7%/1 × TBE ) gel electrophoresis at 23 V for 18 h . The 650-ml cultures of the indicated bacteria grown in TGY to OD600 0 . 9 were harvested by centrifugation , resuspended in 30 ml TGY , and exposed to IR ( 0 °C ) . Irradiated and non-irradiated ( control ) cells were washed and then resuspended in lysis buffer ( 50 mM potassium phosphate buffer [pH 7 . 0] , 0 °C ) . A cell suspension ( ∼4 × 109 cells/ml , 2–4 ml ) was adjusted to 1% ( v/v ) β-mercaptoethanol and passed through a French pressure cell ( 0 °C ) at 20 , 000 lb/in2 , and the lysate centrifuged twice at 12 , 000 × g at 4 °C for 30 min . The protein concentration of a supernatant was determined by the Coomassie ( Bradford ) assay ( BioRad , Hercules , California , United States ) , and the samples were then diluted with lysis buffer/1% β-mercaptoethanol to 20-μg/μl protein , divided into 50-μl aliquots , and stored at −80 °C . The addition of β-mercaptoethanol prior to cell lysis and for storage was recommended in the protocols accompanying the OxyBlot Oxidation Detection Kit ( Chemicon International , Temecula , California , United States ) ( see below ) to prevent the oxidation of proteins during and after cell lysis . Assays for oxidative protein damage ( Figure 3 ) were all conducted in parallel within 3 d of protein extraction . Protein oxidation in freshly prepared extracts was measured using OxyBlot Protein Oxidation Detection Kit ( S7150 ) ( Chemicon International ) , including the indicated molecular-weight markers . The carbonyl groups in the protein side chains were derivatized to 2 , 4-dinitrophenylhydrazone by reaction with 2 , 4-dinitrophenylhydrazine ( DNPH ) for 15 min in 3% ( w/v ) SDS . Western blotting: the DNP-derivatized protein samples were separated by polyacrylamide denaturing gel electrophoresis ( 4%–20% gradient gels; BioRad ) at 195 V for 50 min followed by transferring proteins to a nitrocellulose membrane for 40 min ( BioRad ) . The membranes were incubated with primary antibody , specific to the DNP moiety of the proteins . This step was followed by incubation with a horseradish peroxidase-antibody conjugate directed against the primary antibody ( secondary antibody: goat anti-rabbit IgG ) . The membranes were then treated with chemiluminescent ( SuperSignal ) substrate ( Pierce Biotechnology , Rockford , Illinois , United States ) and imaged by exposure to light sensitive films ( BIOMAX Light Film; Kodak , Rochester , New York , United States ) . O2 and H2O2 concentrations were determined by the Rhodazine D assay ( RDA ) ( CHEMetrics , Calverton , Virginia , United States ) . All Quick-Seal centrifuge tubes containing cells were centrifuged at 4 °C ( 2 , 000 × g , 10 min ) after irradiation . Supernatants were tested by the RDA . The RDA is suitable for measuring 0 . 1–1 . 0 mg/l O2 , and 1–10 × 10−5 M H2O2 . Mn ( III , IV ) are oxidants that will also cause a positive RDA test result . However , Mn ( III , IV ) dioxides are insoluble in water ( circa-neutral pH ) and readily removed from suspension by centrifugation ( 2 , 000 × g , 5 min ) or standing . RDA results indicating 1 mg/l or more O2 or 10 × 10−5 M or greater H2O2 were also tested by the Indigo Carmine assay ( CHEMetrics ) , which yields a blue color suitable for measuring higher concentrations of O2 and H2O2 . Once a test solution had filled an assay vial , the open end was sealed anaerobically with vacuum grease , and the vial was stored in the dark . Mn K-edge XANES spectra were measured [44] in transmission on standard compounds ( MnCl2 , γ-MnOOH , and MnO2 ) and in fluorescence on the D . radiodurans samples . The measurements on γ-MnOOH and MnO2 were made on fine powders spread evenly onto adhesive tape and folded to make samples of appropriate thickness for the transmission experiment . The 1 M MnCl2 solution and D . radiodurans cell suspensions ( OD600 0 . 9 ) were absorbed on paper and cellulose acetate filters , respectively , frozen , and then stored at −80 °C . For XANES measurements , samples were transferred from dry ice ( −78 °C ) to a sample stage consisting of an insulated Peltier stack kept under a helium atmosphere . Such samples were maintained at −14 °C during the analysis . The XANES spectra were measured at the MRCAT [45] ( Materials Research Collaborative Access Team ) sector 10ID beamline at the Advanced Photon Source ( APS ) . The 10ID is an insertion device beamline using APS undulator-A [46] . The undulator was tapered by approximately 2 keV to reduce the variation in incident radiation to less than 15% over the energy range used in these measurements . The incident X-ray beam was the undulator fundamental and was monochromated using the ( 111 ) reflection of a liquid-nitrogen–cooled , double-crystal , silicon monochromator . Higher harmonic content of the incident beam was rejected using a polished float glass mirror . Ionization chambers were used to measure the incident and transmitted intensities , and were filled with 10% nitrogen/90% helium and 100% nitrogen , respectively . The fluorescence spectra were measured with an ionization detector in the Stern-Heald geometry [47] filled with Ar gas , and used a chromium oxide filter of three absorption-lengths thickness to reduce the background signal . The incident X-ray beam was 1 mm square . Linearity tests [48] indicated less than 0 . 04% variation for a 50% decrease in incident X-ray intensity . The XANES data were processed using the Athena program [49] . All data scans were aligned in energy using a reference spectrum derived from metallic manganese . Between five and 20 scans per sample were averaged to improve measurement statistics . The averaged data were normalized [44] by regressing a line to the pre-edge region between approximately 6 , 400–6 , 500 eV . This line was subtracted from the data . A quadratic polynomial was regressed to the region from approximately 6 , 600 eV to the end of the data range . This quadratic polynomial was extrapolated to the edge energy ( 6 , 548 . 5 eV for the samples containing D . radiodurans ) . The normalized data were divided by the value of this polynomial at the edge energy . Notably , the edge positions of XANES spectra of irradiated or non-irradiated D . radiodurans samples were essentially the same as for aqueous MnCl2 solutions . This is consistent with oxygen atoms surrounding Mn ( II ) in D . radiodurans , and with no significant change in Mn valence during irradiation , which would have been observed in cell samples containing 5% or more Mn ( III , IV ) . XRF microprobe analysis measurements [25] were made at beamline 2ID-D at the APS [50] . The 2ID-D is an undulator beamline with Fresnel zone plates focusing optics that produced a focal spot with a FWHM ( full width at half maximum ) spatial resolution of approximately 120 nm for these experiments . A silicon ( 111 ) double-crystal monochromator was used to shine 10-keV photons on the Fresnel zone plates . The sample was rastered through the focal spot of the X-ray beam . For each pixel , the full XRF spectrum between approximately 2 keV and 10 keV was measured using a silicon drift detector . Thus , the distribution of elements between phosphorus and zinc on the periodic table of elements could be measured with 120-nm resolution throughout a cell and its periphery . XRF microprobe measurements were made on D . radiodurans cells grown to OD600 0 . 3 . The cells were deposited on grids as suspensions in TGY liquid medium , which served to help maintain the structure and viability of the cells as they dried . The complete results for two distinct diplococci ( no . 109 and no . 110 ) are presented in Figure S3 . The position coordinates of individual diplococci mounted on formvar-coated gold TEM grids were determined by light microscopy ( LM ) . TEM grids were subsequently placed in the focal plane of the X-ray optic , and the grid markers were used to locate the targeted cells . Data from 2ID-D were initially processed using the program MAPS [51] . The average XRF spectral intensities from 28 adjacent pixels in the upper right-hand corners of XRF images , in areas beyond the boundaries of a cell ( Figure S3 ) , were used to determine the background signal for the elements under investigation . The average background spectrum for an element being mapped was subtracted from all other pixels of the image . Next , a mathematical model representing the original morphology of cells was constructed in approximate likeness to diplococci , which resemble adjoined truncated spheres . Within each pixel , the intensity of an element's XRF signal was weighted by the average thickness of the diplococcus subtended by a respective pixel . The contour lines are straight-line segments between points of equal intensity , in which linear interpolation of the actual data was used to determine the locations of the contour values . The contours were generated using the algorithm Gnuplot ( http://www . gnuplot . info ) . In combination , this approach limited the effect of cell thickness on the elemental distribution maps , and facilitated correlation between electron-dense granules ( EDGs ) and Mn hotspots . Element distributions are presented as mass parts per million ( ppm ) . Whole cells deposited on formvar-coated gold TEM grids and analyzed by XRF microprobe analysis were subsequently imaged in a FEI Tecnai TF30 transmission electron microscope equipped with a field-emission gun and operated at an accelerating voltage of 300 kV . An objective aperture with a diameter of 40 μm was selected to optimize contrast due to differences in elastic scattering across the specimen . Bright-field electron micrographs were recorded using a Gatan Ultrascan Model 894 cooled CCD camera containing four mega-pixels and operated with a four-port parallel read-out . Images were analyzed using Gatan Digital Micrograph software ( version 3 . 4 ) . Pixel intensities were transformed to a logarithmic scale to visualize the EDGs ( ∼ 0 . 2 μm in diameter ) in cells . LM was used to identify the position coordinates of discrete diplococci on formvar-coated gold TEM grids before XRF microprobe analysis ( Figure S3 ) ; gold grids were supported by glass slides during LM . LM images were captured using a Leica DM RXA epi-fluorescent microscope ( Leica , Wetzlar , Germany ) . Differential interference contrast images were recorded using a Scion Corporation ( Frederick , Maryland , United States ) CCD camera and analyzed with xyzGrabber . | One original goal of radiobiology was to explain why cells are so sensitive to ionizing radiation ( IR ) . Early studies in bacteria incriminated DNA as the principal radiosensitive target , an assertion that remains central to modern radiation toxicity models . More recently , the emphasis has shifted to understanding why bacteria such as Deinococcus radiodurans are extremely resistant to IR , by focusing on DNA repair systems expressed during recovery from high doses of IR . Unfortunately , as key features of DNA-centric hypotheses of extreme resistance have grown weaker , the study of alternative cellular targets has lagged far behind , mostly because of their relative biological complexity . Recent studies have shown that extreme levels of bacterial IR resistance correlate with high intracellular Mn ( II ) concentrations , and resistant and sensitive bacteria are equally susceptible to IR-induced DNA damage . The current work establishes a mechanistic link between Mn ( II ) and protection of proteins from radiation damage . In contrast to resistant bacteria , naturally sensitive bacteria are highly susceptible to IR-induced protein oxidation . We propose that sensitive bacteria sustain lethal levels of protein damage at radiation doses that elicit relatively little DNA damage , and that extreme resistance in bacteria is dependent on protein protection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"cell",
"biology",
"physiology",
"microbiology",
"chemical",
"biology",
"molecular",
"biology",
"eubacteria"
] | 2007 | Protein Oxidation Implicated as the Primary Determinant of Bacterial Radioresistance |
Intron number varies considerably among genomes , but despite their fundamental importance , the mutational mechanisms and evolutionary processes underlying the expansion of intron number remain unknown . Here we show that Drosophila , in contrast to most eukaryotic lineages , is still undergoing a dramatic rate of intron gain . These novel introns carry significantly weaker splice sites that may impede their identification by the spliceosome . Novel introns are more likely to encode a premature termination codon ( PTC ) , indicating that nonsense-mediated decay ( NMD ) functions as a backup for weak splicing of new introns . Our data suggest that new introns originate when genomic insertions with weak splice sites are hidden from selection by NMD . This mechanism reduces the sequence requirement imposed on novel introns and implies that the capacity of the spliceosome to recognize weak splice sites was a prerequisite for intron gain during eukaryotic evolution .
Intron number is highly variable among eukaryotes , ranging from about a dozen in some fungi to more than 100 , 000 in the human genome . Comparative genomics across broad phylogenetic distances have identified the importance of both intron gain and loss to the establishment of this variation [1] . In particular for a number of lineages , including Drosophila [2] , Caenorhabditis [3] and some isolated vertebrate lineages [4] , a considerable number of intron gains have been described . While there is a general agreement that the very first spliceosomal introns arose from the degeneration of self-splicing group II introns [5] , [6] , their complete absence from genomes that have undergone intron gain strongly suggests alternative mechanism ( s ) are at work . While several mechanism with varying levels of empirical support have been proposed over the last 30 years , there is still strong uncertainty over whether any existing model can explain the observed and predicted rates of intron gain throughout eukaryote evolution [7] . A satisfactory model must address the mutational mechanism that allows a intron to colonise a novel position and the evolutionary process that facilitates the fixation of this new allele within a population . An accounting of both mechanism and evolution should give insight into why the rate of intron gain is so variable between species . Irrespective of the mutational mechanism , it is apparent that any new intron will require a number of key motifs including the 5′ and 3′ splice sites , and a set of auxiliary signals including the branch point and splicing enhancer and suppressor motifs [8] , [9] . The failure to correctly identify an intron may either lead to stochastic alternative splicing or intron retention , both of which have deleterious consequences . This predicament is overcome if the newly inserted intron arrives fully functional . The only mechanism capable of generating a fully formed novel intron is reverse splicing [10] , [11] , in which an existing intron propagates into a new position , but this process is both extremely rare and inconsistent with the characteristics observed of novel introns [2] . The alternative is that novel introns develop gradually via the optimisation of previously non-intronic sequence . Examples include the intronisation of coding sequences [3] , intron gain between paralogs of multi-copy gene families [4] , the splicing of an Alu element [12] , after internal gene duplication ( including tandem duplication ) [13] and after the insertion of new sequence of unknown origin [14] . In this study , we have investigated this alternative model in which novel introns are not required to be fully functional , relying instead on a back up mechanism of transcript quality control for incorrectly spliced introns [15] . In recent years it has become evident that the cell invests heavily in the identification of premature termination codons ( PTCs ) via the Nonsense Mediated Decay ( NMD ) pathway [16] , [17] . NMD acts during the preliminary round of translation to identify in-frame stop codons and classify them as either genuine or premature . The use of incorrect splice sites or intron retention are a ready source of such premature termination codons ( PTCs ) and will invoke the NMD dependent destruction of the transcript . Using comparative genomics of nine Drosophila species , we show that novel introns have weaker splice sites and carry more stop codons than conserved introns . We propose that NMD may play an important role during the establishment of novel introns within a population , and in support of this we identified a significant deficiency of novel introns that would remain invisible to the NMD pathway upon intron retention .
Here we have identified 307 novel introns amongst 284 genes across nine Drosophila genomes ( Figure S1 ) , presenting the most comprehensive set of novel introns to date . Our approach also detected 803 intron loss events amongst 595 genes , including 49 genes that have undergone both intron gain and loss ( Dataset S1 ) . These events show a strong heterogeneity across the Drosophila phylogeny , with several lineages being hot-spots of intron turnover ( Figure 1 and Figure S2 ) . We observe the highest rate of intron gain reported thus far , 2 . 8 intron gains/gene/Bya ( 109 ) years in the melanogaster subgroup , being 6× greater than previously reported for Drosophila [2] and 4× greater than the next highest reported rate ( occurring in yeast ) [1] , [18] . Interestingly , this rate is still higher than the range of estimates required to have generated the intron-rich eumetazoan genome ( 0 . 99–2 . 39 gains/gene/Bya years ) [1] , [19] . In sharp contrast , several other Drosophila lineages have experienced far less intron gain . D . virilis underwent only 0 . 0022 intron gains/gene/Bya years and since the split between D . melanogaster and D . yakuba 10 million years ago not a single intron gain could be identified , demonstrating that the rate of intron gain may vary over orders of magnitude between closely related species . The previously proposed mechanisms of intron gain assume that new intronic sequence originates from elsewhere in the genome ( reverse splicing [11] , [20] or mobile elements [10] , [12] , [21] ) , or is derived from the endogenous genomic location ( tandem duplication [22]–[25] or intronisation [3] , [7] ) . Despite a rigorous search ( Text S1 ) we could not identify an homologous parental origin for any novel intron elsewhere within the respective genomes , consistent with other studies [2] , [26] . A manual inspection of the sequence flanking each novel intron identified a single event reminiscent of tandem duplication . The Bap170 gene in D . pseudoobscura has undergone a gain of 218 bp , of which only 206 bp are spliced out , revealing an imperfect 8 amino acid repeat 5′ and 3′ of this novel intron ( Figure S3 ) . While in final stage of preparing this manuscript Li et al . , ( 2009 ) reported that several novel introns in Daphnia are flanked by short direct repeats [14] . They insightfully suggest this may represent the signature of nonhomologous end joining ( NHEJ ) after uneven double-stranded breaks ( DSBs ) , a process known to generate insertions flanked by direct repeats [27] . In consideration of this , we note that the duplication observed here may also be explained by a direct repeat flanking sequence of unknown origin . A manual inspection of dotplots identified 6 further examples in which direct repeats of length at least 8 bp overlapping the splice sites of a novel intron ( Figures S4 , S5 , S6 , S7 , S8 , S9 ) in support of the finding of Li et al . , ( 2009 ) . Reasoning that changes to the length of the coding sequence directly flanking a novel intron , as observed for Bap170 , may give further insight into the mechanism of intron gain , we checked all 307 novel introns for alterations to the coding sequence that would indicate either the loss or gain of adjacent amino acids . Novel introns did not alter the ancestral coding sequence in 87% ( 267/307 ) of the cases . The remaining 13% ( 40/307 ) modified the adjacent coding sequence by only 1–3 amino acids ( in 3 cases there was a gain of 4 or 5 amino acids along next to the new intron ) . This observation is inconsistent with the intronisation model of intron gain [3] , [7] which requires the conversion of exonic sequence into an intron , hence reducing the coding sequence by the size of the new intron . A manual inspection of these 40 coding sequence-changing novel introns identified a small number of cases that can be explained via the conversion of low complexity amino acid sequence into an intron ( Figure 2 and Figure S10 ) . The novel intron within gene CG42594 has arisen from a rapidly evolving low complexity region including poly-Q sequence . Species lacking this intron show a highly variable sequence of amino acids at this position , with length differences of up to 18 amino acids . In the ancestor of D . melanogaster and D . ananassae this low complexity amino acid sequence was converted into an intron , stabilising the flanking protein sequence , while freeing the new intronic sequence of length constraint . This indicates that the expansion of protein sequences can generate novel introns . Indels account for the majority of sequence variation between Drosophila species ( 3 . 2% of variable nucleotides vs . 1 . 8% for SNPs [28] ) making them a significant contributor to both coding and non coding length evolution . Previous work focused on the mechanism underlying relatively short insertions ( <15 bp ) , therefore , to access the possible contribution of exonic insertions to intron gain we identified insertions long enough to generate a novel intron ( >44 bp in Drosophila ) . This revealed 180 insertions ( Dataset S1 ) , the largest being an insertion of 165 amino acids within the XNP gene of D . pseudoobscura . This demonstrates the plasticity of protein length and establishes large insertions within the protein coding sequence of Drosophila as a viable source of novel intronic sequence . We reason , that a much larger number of exonic insertions occur over evolutionary time providing the raw genetic variation for the gain of novel introns . The model that novel introns arise from a subset of “random” insertions within coding regions ( or indeed UTR sequences ) predicts that new introns are unlikely to arise with full strength splice sites . We observe that novel introns do in fact have weaker splice sites , with significantly reduced usage of the “strong” consensus motif at both the 5′ and 3′ splice site ( Figure 3A and Figure S11 ) . Furthermore , novel introns use a more diverse set of rare 5′ motifs than expected ( Figure S11 and Dataset S1 ) . Of course , weak , rare or atypical 5′ splice sites have lower affinity to the U1 snRNP of the spliceosome [29] which , all else being equal , leads to less efficient splicing [30] , [31] . This poses a conundrum; if the mutational mechanism that generates novel introns leaves them vulnerable to suboptimal splicing , why do such novel introns rise to fixation within a population ? We propose that the solution lies in the action of NMD . Retention of 3n+1 and 3n+2 introns is expected to induce NMD due to the introduction of a frame-shift , but introns of length 3n require an in-frame PTC or they will remain invisible to the NMD pathway . Because of this , we reason that the failure to splice a new 3n insertion maybe deleterious , hence we predicted that novel 3n introns are more likely to encode a PTC as a backup mechanism for incomplete splicing . As the expectation for PTC occurrence is proportional to intron length , we fitted a logistic regression , modelling intron length , intron phase and a combined main effect of 3n class ( 3n vs . 3n+1 and 3n+2 ) and whether an intron is novel ( n = 307 ) or conserved ( n = 8 , 810 ) ( Text S1 ) . Despite its simplicity , our model was highly significant ( P<0 . 0001 ) and explained 24% of the variation in the occurrence of stop codons among introns . Interestingly , most of the variation was explained by phase ( Wald χ2 = 331 . 5 , P<0 . 0001 ) and not intron length ( Wald χ2 = 174 . 2 , P<0 . 0001 ) . Phase 2 introns encode significantly more in-frame PTCs than either phase 0 and 1 due to the sequence requirements of the 5′ splice site . The canonical 5′ splice site GT ( A/G ) A restricts the first full potential codon of a phase 2 intron to either the TAA Ochre or TGA Opal stop codon . Only a minority of introns with non-canonical splice sites escape this constraint . Our analysis indicates that selection acts against introns that are invisible to the NMD pathway ( if they undergo intron retention ) leading to a deficiency of 3n PTC-free introns across the genome , as previously reported [32] ( Figure 3B ) . This verifies in Drosophila that NMD carries a significant load caused by the weak splicing of introns [16] . We also observe this deficit of 3n PTC-free introns within the 307 novel introns . Interestingly , we find that this effect is significantly stronger among novel introns than among conserved introns ( Odds ratio of 3 . 027 for novel vs . 1 . 646 for conserved ) , supporting the central role of NMD in the establishment of newly inserted sequence as novel introns . Here we have shown that while the expansion of amino acid repeats within exons can generate novel introns , nevertheless , the sequence origin for the vast majority remains unknown . This observation is inconsistent with previously suggested mechanisms of intron gain , but supported by the recent study of novel introns within Daphnia [14] . We have demonstrated that novel introns in Drosophila use weaker splice sites and are deficient for 3n PTC-free introns . Therefore , our evidence suggests that the establishment of these new sequences as introns is facilitated by NMD . Therefore , we propose a new model of intron gain ( Figure 4 ) , in which mutational mechanisms generate insertions that already carry the minimal requirements for correct , but not necessarily strong splicing . Cytoplasmic NMD is expected to degrade any unspliced transcript , leaving a proportion with the correct coding sequence . Conditional on adequate expression levels , this will shelter the new intron from selection allowing it to segregate within the population as a neutral polymorphism . Importantly , NMD allows new introns to utilise a more degenerate set of splice sites , thereby increasing the likelihood that any new sequence may become captured by a novel intron . This model makes several predictions: First , novel introns are not required to pass through a protein coding intermediate stage ( as would be expected from the intronisation of existing exonic sequence ) and therefore , should not show codon usage bias . We observed no correlation between the “codon” usage of novel introns and the expected codon usage for Drosophila genes ( Spearman Correlation Coefficient 0 . 01983 , P = 0 . 8764 ) ( Figure S12 and Text S1 ) . Second , in general , introns with weaker splice sites are expected to suffer higher rates of failed splicing ( intron retention or exon skipping ) , but we observe less intron retention among novel introns ( 2 . 6% ) compared to conserved introns ( 5 . 3% ) . This is consistent with our expectation that via the action of NMD these transcripts are removed . The “faux 3′ UTR” model suggests that PTC recognition depends on the distance to the downstream polyA tail [33] , [34] . This makes NMD more potent towards the 5′ end of the transcript , leading to a third prediction; the establishment of novel introns should also be more efficient towards the 5′ . As expected , we identified a strong and highly significant 5′ bias for novel introns ( χ2 = 26 . 063 , P<0 . 001 ) ( Figure S13 and Text S1 ) in support of previous work [2] . NMD is more effective towards the 5′ as a PTC located towards the 3′ is more likely to be recognised as a canonical stop codon [7] . Hence , the involvement of NMD in the establishment of novel introns can explain the thus far enigmatic 5′ bias observed within a number of species [2] , [7] , [35] . The 803 lost introns reported here show no positional bias ( χ2 = 1 . 309 , P = 0 . 2526 ) , consistent with previous reports [2] , [18] , [36] . In addition to 3′ UTR length the exon junction complex can invoke NMD in mammals . In effect this allows the recognition of PTCs in close proximity to the polyA tail , enhancing the effectiveness of NMD towards the 3′ of a transcript . Testing the influence of this on the distribution of novel introns is difficult due to their scarcity , but we note that mammalian genomes do not show the 5′ bias among all intron seen in Drosophila [35] . A significant question remains why does the rate of intron gain vary so much between closely related species ? While differences in the action and potency of NMD are likely to exist between highly divergent taxa , we do not expect much variation on the fine scale of the Drosophila phylogeny . In contrast , the mutational processes that generate repeat expansions , tandem duplications [13] , insertions of unknown origin [37] and DSBs are known to vary greatly between both closely and distantly related species . Differences in these underlying mechanisms will generate species specific variation upon which our proposed mechanism of intron fixation may act . This offers a possible explanation for the variation in intron gain rates observed here and over longer periods of eukaryotic evolution .
Our approach to studying intron evolution is based on identifying gene orthologs across the Drosophila clade , predicting gene structure with GeneWise and using Dollo Parsimony to infer intron gain and loss events ( Figure S1 ) . We identified orthologous genes using the D . melanogaster ( release 4 . 3 ) gene set as the basis of a best-bidirectional-blast-hit approach in the 11 other sequenced Drosophila species , namely; D . erecta , D . yakuba , D . ananassae , D . pseudoobscura , D . willistoni , D . virilis , D . mojavensis and D . grimshawi ( obtained from http://rana . lbl . gov/drosophila/ ) . We excluded D . sechellia , D . simulans and D . persimilis because of low sequence coverage [38] . We acknowledge that a bidirectional-blast approach carries limitations but given our subsequent validation of intron turnover events feel this method was suitable . High-scoring segment pairs ( HSPs ) were identified via forward tblastx with default parameters followed by reverse tblastx using sequence cropped on either side of the best hit equivalent to the length of the corresponding gene in D . melanogaster . We considered the HSPs to be orthologous when the reverse blast identified only the parental gene in D . melanogaster . Exon-intron structure of orthologous genes was generated by submitting to GeneWise [39] ( 2193 algorithm ) the longest amino acid isoform of each D . melanogaster gene together with 100kb of nucleotide sequence flanking the corresponding orthologous hit . We excluded any gene with a frameshift mutation ( either real or due to sequencing errors ) . Intron gain and loss events were predicted using the Malin java application [40] . The dense phylogeny of sequenced Drosophila genomes increases the power of Dollo Parsimony to accurately infer intron gain events , reducing the advantages of maximum likely methods [41] . Along two branches of the phylogeny ( leading to D . willistoni and D . grimshawi ) Dollo Parsimony remains sensitive to multiple losses being inferred as intron gain , but given the active debate about the best methods to infer intron turnover [42] we feel our approach and extensive downstream validation have proved reliable . As our approach relies on de novo gene structure prediction via GeneWise it is sensitive to false positive and false negative intron prediction in other species . This problem was avoided in a previous study by considering only introns present in the well annotated D . melanogaster lineage [2] . Our approach takes full advantage of the multiple sequences genomes to find intron gain events outside of D . melanogaster , but required extensive validation to overcome the several limitations of GeneWise ( detailed in Text S1 and Dataset S1 ) . This approach generated a high confidence set of 3 , 593 fully annotated orthologous genes ( containing 8 , 810 introns ) across nine Drosophila species , allowing us to identify intron gain and loss events across 40Mys of Drosophila evolution . Our approach is based on the amino acid sequence in D . melanogaster and is therefore not able to predict UTR introns . After this we still expected our data set to contain false positives ( predicted novel introns that are not really introns ) and false negatives ( real introns that have been missed ) . Our experimental and informatic methods for their identification and exclusion are detailed in the Text S1 . Novel intron sequences and gene , protein and intronic sequences for our orthologous gene set are available for download at http://i122server . vu-wien . ac . at/Drosophila_annotation/ . As per previous studies [43] , [44] , we used the percentage of introns with the consensus 5′ splice site GT ( A/G ) AGT ( position +1 to +6 ) as a measure of the splice site strength within each class of introns . To confirm that novel introns use this motif significantly less than all introns we resampled ( bootstrap with replacement ) 307 introns from the population of 50 , 836 D . melanogaster introns 10 , 000 times ( Figure S11A ) . The top and bottom 2 . 5% of samples gave the 95% confidence intervals on the observed percentages for all introns . The observed percentage of novel introns fell outside these confidence intervals establishing significance . Resampling ( 307 from 307 , with replacement ) from novel introns ( black bars in Figure S11A and S11B ) gives an indication of the variance within novel introns , but is not actually required to establish the significance between all and novel . We repeated this approach for the CAG motif at −3 to −1 of the 3′ splice site ( Figure S11B ) . To show that novel introns use a more diverse set of rare/weak motifs at the 5′ we used the same bootstrap data from above and counted the number of different motifs present in each sample ( Figure S11C ) . | The surprising observation 30 years ago that genes are interrupted by non-coding introns changed our view of gene architecture . Intron number varies dramatically among species; ranging from nine introns/gene in humans to less than one in some simple eukyarotes . Here we ask where new introns come from and how they are maintained in a population . We find that novel introns do not arise from pre-existing introns , although the mechanisms that generate novel introns remain unclear . We also show that novel introns carry only weak signals for their identification and removal , and therefore depend on nonsense-mediated decay ( NMD ) . NMD maintains RNA quality control by degrading transcripts that have not been spliced properly . We propose that NMD shelters novel introns from natural selection . This increases the likelihood that a novel intron will rise in frequency and be maintained within a population , thus increasing the rate of intron gain . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/comparative",
"genomics"
] | 2010 | Nonsense-Mediated Decay Enables Intron Gain in Drosophila |
Clusters of differentiated cells contributing to organ structures retain the potential to re-enter the cell cycle and replace cells lost during development or upon damage . To do so , they must be designated spatially and respond to proper activation cues . Here we show that in the case of Drosophila differentiated larval tracheal cells , progenitor potential is conferred by the spatially restricted activity of the Snoo transcription cofactor . Furthermore , Dpp signalling regulated by endocrine hormonal cues provides the temporal trigger for their activation . Finally , we elucidate the genetic network elicited by Snoo and Dpp activity . These results illustrate a regulatory mechanism that translates intrinsic potential and extrinsic cues into the facultative stem cell features of differentiated progenitors .
Facultative stem cells have been defined as a particular class of differentiated cells that contribute to the structure and function of well-developed organs but remain multipotent; thus , upon damage due to either regular usage or injury they can proliferate and their progeny acquire the identities of different cell types that comprise the organ . While this property is fundamental to ensuring organ development and homeostasis , we still lack a detailed understanding of how these cells are set apart and how they express their progenitor features . We address this issue by the study of a group of progenitor cells in Drosophila with the features of facultative stem cells , namely the Differentiated Adult Progenitors ( DAP ) cells of the larval trachea [1] . Like most Drosophila larval cells , larval tracheal cells are polyploid and die at metamorphosis without contributing to the adult trachea [2] . However , among the larval tracheal cells , some cells escape the endocycle and by doing so acquire the features of progenitor cells of the adult trachea [3] . These cells remain quiescent during larval growth , reactivate their proliferation at the last larval stage and give rise to the different cell types of the adult tracheal network during metamorphosis [1 , 4–6] . DAP cells belong to the dorsal trunks ( DT ) , the main tracheal branches in the larvae and are specific to the second tracheal metamere ( Tr2 ) . The difference between the DT cells in Tr2 and those of the DT in other metameres is established by homeotic genes that exclude fzr expression from the DT cells in TR2 [3 , 6] . fzr encodes the Drosophila homolog of the CDH1 protein of the E3 ubiquitin ligase APC/C and is the gene associated with endocycling in Drosophila [7–9] . While exclusion of fzr expression from the DT cells of TR2 sets a permissive state that allows for the deployment of the genetic program associated with the adult progenitor fate , it remains unclear what is the trigger for said progenitor program . Here we identify the Snoo transcriptional cofactor as the executor of the adult progenitor program , which is downregulated by Fzr in metameres other than TR2 . However , while Snoo and Fzr are both spatially regulated , they do not show any temporal specificity and thus cannot account for the temporal control of DAP cell proliferation at the third larval stage ( L3 ) . In this regard , we show here that mitotic activation of DAP cells as well as expression of progenitor features also requires Decapentaplegic ( Dpp ) /Tgf-β signalling , which is not active in the DT cells at L2 but is activated at L3 . Further , we show Dpp signalling to be under the positive control of Ecdysone and the negative control of Juvenile Hormone thus providing the readout for the transition from L2 to L3 . Finally , we report that together , Snoo and Dpp regulate the mitotic factor String ( Stg ) /cdc-25 as well as Broad ( br ) , a key regulator of metamorphosis to promote the proliferation and progenitor behaviour of DAP cells . In sum , we show how the spatial restriction of Snoo protein and timed activation of Dpp provides sufficient instruction for differentiated tracheal cells to fulfil their progenitor potential .
The genetic program associated with adult progenitor cell fate is suppressed by Fzr in the DT cells of metameres other than TR2 probably by downregulation of one or more specific transcription factors . Among the possible targets for Fzr mediated degradation , the mammalian Ski/Sno oncogenes seemed an interesting candidate [10] as they are both a target of CDH1 and a regulator of cell proliferation . We thus reasoned that the Drosophila homologue , Snoo , might play a role in DAP cell behaviour [10] . Like its mammalian counterparts , Snoo contains a highly conserved d-box motif , which likely results in targeting for degradation by the APC/C [11] . Consistent with the hypothesis that Snoo is responsible for activating the adult progenitor genetic program , forced overexpression of Snoo in the entire trachea results in ectopic expression of Headcase ( Hdc ) , a marker for adult progenitor cell fate [3] , whose expression in the wild-type is restricted to the DT cells in TR2 ( Fig 1A and 1B ) Conversely , RNAi mediated knockdown of Snoo causes loss of Hdc expression in the DT cells of TR2 ( Fig 1C ) . Interestingly , some nuclei in the Tr2 DT of snoo ( RNAi ) trachea appear larger than their WT counterparts . This is likely due to an incomplete penetrance upon Snoo RNAi in the loss of stg expression , which is associated with the mitotic potential of the cells; thus these cells are likely larger due to the fact that they are at 4N after having gone through an S-phase and are paused at G2 [3 , 6] . Transcription of the snoo gene occurs all over the DT larval tracheal cells including DAP cells as assessed by a construct driving lacZ expression under the control of the snoo promoter ( Fig 1E ) [12] . To examine whether Fzr promotes the degradation of the Snoo protein , and since we could not rely on any available antibody , and having failed in producing a Snoo specific antibody , we made use of a transgenic strain carrying a GFP tagged form of Snoo under the control of UAS promoter sequence [13] . Following a pulse of btlGal4 mediated expression of the tagged form of Snoo ( see mat and methods ) , we recovered tracheae that accumulated Snoo::GFP in the DT cells in the TR2 metamere , which do not express fzr . However , we observed very low levels of Snoo::GFP in the DT cells posterior to the TR2 metamere , which do express fzr ( Fig 1F ) . Conversely , continuous expression of the same construct leads to Snoo::GFP accumulation along the DT cells of all metameres ( Fig 1G and 1H ) suggesting that under these circumstances endogenous Fzr protein cannot degrade the high amounts of induced Snoo::GFP protein , thus accounting for the general expression of adult progenitor markers ( see above ) . Finally and consistent with these observations , accumulation of the DAP marker ( Hdc ) throughout the DT cells of all metameres upon forced overexpression of Snoo can be reversed by coexpression of Fzr ( Fig 1D and 1I ) . All together these results identify Snoo as a transcription factor triggering the DAP genetic program and being antagonized by Fzr . Upon tracheal overexpression of Snoo we also observe DT cell division ( as indicated by pH3 staining ) in the metameres where DT cells normally enter into endocycle and do not divide ( S1A and S1B Fig ) . Since Fzr negatively regulates activity of the stg locus [3] , thus preventing cell division , we reasoned that , similar to the case for Hdc , Snoo may also activate stg expression . In this way , Fzr mediated degradation of Snoo would account for the negative effect of Fzr on stg expression . Consistent with this hypothesis , stg expression in the DT cells in Tr2 is lost following RNAi-mediated knockdown of Snoo ( Fig 1J and 1K ) while Stg-GFP accumulates ectopically in the DT cells posterior to the TR2 metamere upon overexpression of snoo ( Fig 1L ) . Thus , Snoo appears to be necessary and sufficient to trigger both proliferation and expression of adult progenitor cell markers in the DT cells . Expression of specific markers and onset of proliferation of DAP cells occurs at mid third larval instar ( L3 ) , the last larval stage before pupariation and metamorphosis . Thus , because snoo is transcribed at earlier larval stages ( Fig 1E ) snoo expression cannot be responsible for providing the temporal cue for L3 activation of the DAP cell program , suggesting that a temporally regulated trigger must cooperate with Snoo . To investigate which this temporally regulated trigger might be we took into consideration previous reports on the interaction between Ski/Sno proteins and TGF-beta signalling [12–14] . Thus , we decided to investigate the role of the Dpp pathway , the Drosophila homolog of TGF-beta , in DAP cell proliferation . First , we analysed the timing of Dpp pathway activity by the use of a GFP tagged version of Daughters against dpp ( Dad ) , a reporter of Dpp signalling [15] . At Early L3 , we do not observe Dad::GFP expression in the DT cells ( Fig 2A ) , where it is only detected by mid L3 ( Fig 2B ) , the time of mitotic activation of DAP cells [3] . This pattern mimics that of the phosphorylated form of the R-SMAD , Mothers against dpp ( Mad ) ( S2A–S2C Fig ) . It should be noted that while activation of Dpp signalling during L3 in the wild-type trachea initiates at the anterior metameres it eventually occurs along the DT cells of all the metameres . Furthermore , we find the mitotic potential of DAP cells to be impaired following RNAi mediated knockdown of the DPP receptor Thickveins ( Tkv ) as evidence by loss mitotic marker pH3 ( Fig 2C , 2D and 2I ) as well as loss of expression of stg ( Fig 2E and 2F ) . The abrogation of mitosis via Tkv knockdown is consistent with the recently reported role of Dpp signalling in promoting remodelling in the larval trachea [16] . Remarkably , Dpp signalling is not only required to promote cell proliferation but also for the DAP genetic program as we also observe loss of Hdc expression following RNAi mediated knockdown of Tkv ( Fig 2G and 2H ) and Mad ( S2H Fig ) . Consistent with the Dpp pathway providing a temporal cue for DAP cell proliferation , we observe extra Tr2 DT cells in molting L2 larvae upon precocious Dpp signalling by means of the constitutively activated receptor encoded by the tkvQ253D allele [17] ( Fig 2J , 2K and 2L ) . Precocious activation of the Dpp pathway also triggers an earlier activation of Hdc as well ( Figs 2M , S2D–S2G ) . As Dpp signalling by L3 occurs along the DT cells of all the metameres it seems therefore not to be responsible for the spatial restriction of the DAP cell genetic program to the DT cells of Tr2 . Consistently , ubiquitous tracheal expression of the allele encoding the constitutively activated receptor tkvQ253D does not induce either cell division or ectopic hdc expression outside the Tr2 metamere ( Fig 2J , 2M and 2N ) . All together , these results indicate that Dpp signalling provides the temporal trigger for the activation of the adult progenitor cell markers and for their proliferation . We then addressed how Dpp signalling might be so timely regulated and found that expression of Dpp itself as inferred via dpp-gal4<UAS-gfp [18 , 19] ( Fig 3A and 3B , S3A–S3F Fig ) , the Dpp receptor Tkv as shown by GFP tagged Tkv protein [20] ( Fig 3C and 3D ) and the downstream activator Mad as shown by GFP tagged Mad protein [21] ( Fig 3E and 3F ) in the DT cells is turned on at mid L3 . As timely activation of DAP cells requires hormonal input via Ecdysone [3] , it is likely that Dpp and Tkv expression is itself dependent upon ecdysone signalling . Indeed , genetic ablation of the Ecdysone producing Prothoracic Gland ( material and methods and [3] during the third larval instar abrogates the activation of the dpp pathway ( Fig 3G and 3H ) . However there are several cycles of ecdysone activation during larval life , with ecdysone levels peaking at each molt , whereas Dpp signalling is active in the DT cells only after the molt from L2 to L3 , suggesting the activity of another temporal regulatory input . We reasoned such an additional input could be the Juvenile Hormone ( JH ) . Levels of JH stay at high levels until the third larval stage and then drop to very low levels [22] . In the cockroach Blatella , this drop in JH levels is required for metamorphosis to be triggered by a peak of Ecdysone [23] . Similarly , Dpp , Mad and Tkv expression in the DT cells might be under the positive control of ecdysone and the negative control of JH . Due to complexity of JH signaling , it was not feasible to genetically manipulate the JH pathway [24] . We instead chose to prevent the reduction in JH signalling by exposing larvae to the JH agonist methoprene [25] . Consistent with JH regulation of Dpp signalling , we observed not only reduced signalling activity of DPP , as seen by the Dad::GFP reporter , but also a reduction of cell division as well as loss of Hdc expression in Tr2 DT DAP cells ( Fig 3I–3N ) . While the above results show that Dpp signalling is required for hdc expression ( Fig 2H ) , an analysis with the matscan application [26] did not predict Mad binding sites in the hdc promoter region , suggesting that hdc might not be a direct target of the Dpp pathway . Instead , our predictions showed several binding sites for Mad in the br locus and for Br in the hdc locus . The br gene is a well-known target of ecdysone signalling [27] and is activated in the DT cells of Tr2 at L3 [3] . We thus examined 1 ) whether activation of hdc expression by Dpp signalling might be mediated by br . and 2 ) whether activation of br expression by ecdysone might be mediated by Dpp signalling . Consistent with hdc expression being dependent on Br function , RNAi mediated knockdown of Br results in the loss of Hdc in the DT cells of the TR2 metamere ( Fig 4A and 4B ) . To further analyse the role of br in hdc regulation , we explored whether br activity was sufficient to activate hdc expression in all the DT cells . br encodes four isoforms ( Z1-Z4 ) required for various cellular processes during metamorphosis [28 , 29] . We forced the expression of each of distinct isoform and found that the Z2 isoform of Br was sufficient to activate hdc in the rest of the trachea ( Fig 4C ) . Altogether , these results place hdc expression downstream of br . Finally , and consistent with br expression being a target for Dpp signalling , RNAi mediated knockdown of Mad resulted in loss of br-Z2 expression in DAP cells ( Fig 4D and 4E ) . In addition , we find the expression of the br-Z2 isoform in the DT cells of TR2 to be under the control of Snoo as this accumulation is lost upon RNAi mediated knockdown of Snoo ( Fig 4D and 4F ) . These results indicate that , as is also the case for stg expression , br expression integrates the spatial and temporal triggers from Fzr/Snoo and ecdysone/Dpp signalling respectively . Interestingly , RNAi mediated knockdown of Br did not affect either DT cell division or stg expression ( S4A and S4B Fig ) , which suggests a split in the activation of the programs leading to DAP cells specification and proliferation downstream of Snoo/Dpp . Finally , we want to note that the requirement of Mad for br and stg expression does not imply that the Dpp pathway solely mediates all the effect of ecdysone in these genes . In this regard , we observed overlap at the promoter regions of br and stg when comparing ChIP-Seq data of EcR binding produced from whole animals at prepupal stages with predicted Mad binding sites using the Gbrowse function of modENCODE [30] . These observations suggest the combined activity of Ecdysone receptor as well as Mad/Snoo might ensure a robust activation of the DAP cell genetic program . The main distinction between undifferentiated progenitors and “facultative stem cells” is the continuous growth required for normal cellular turnover of the first vs activation of the later in singular conditions such as injury [31 , 32] . Thus , facultative stem cells which functionally contribute to an organ , yet retain their multipotency , must ignore the signals that mediate the continual cellular turnover that maintains tissue homeostasis . Only upon specific events should they activate and realize their potential as progenitors . In the fly , continuous growth of undifferentiated progenitors stimulated by hormonal signalling is exemplified by the growth of the imaginal discs throughout larval life . On the other hand , the precise triggering of facultative stem cell growth can be exemplified by the activation of tracheal DAP cells . Where the growth of imaginal disc cells is continuous , activation of DAP cells requires additional distinct spatial and temporal cues . The findings here reported show how spatial restriction of potential via Hox mediated activation of Fzr and degradation of Snoo ensures that the temporal cue of Ecdysone in the absence of the juvenile hormone and mediated by activation of Dpp signalling will be interpreted by the correct population of cells in response to the correct organismal context ( Fig 4G ) . These results illustrate a mechanism by which factors acting in differentiated cells to maintain the potential for pluripotency cooperate with the factors triggering their transition from quiescence to proliferation in order to carry out their function in regenerative growth; this may provide a general framework for studying facultative stem cells and lend insight into what may be missing in similar cells that are incapable of such potential and/or to respond to the temporal trigger to activate their potential .
The following flies were obtained from the Bloomington Stock Center: UAS-tkvRNAi , UAS-snooRNAi , BrZ2::GFP , Mad::GFP , UAS-brZ1 , UAS-brZ2 , UAS-brZ3 , and UAS-brZ4 . UAS-madRNAi and UAS-brRNAi , were obtained from the VDRC . Stg::GFP ( YD0246 ) was obtained from the Yale Flytrap Project . The following strains were given by: UAS-snoo; UAS-fzr ( Rosa Barrio ) , UAS-snoo ( Rosa Barrio ) , UAS-snoo::GFP ( Jose de Celis ) , the R6 . 4 stg-lacZ ( Lehman et al , 1999 ) , UAS-tkvQ253D ( Lecuit et al 1996 ) , Tkv::GFP ( Hsiung et al , 2005 ) , dpp-Gal4<UAS-GFP ( María Domínguez ) , snoo-lacZ ( SH1402 ) ( Stuart Newfeld ) and Dad::n::GFP ( Enrique Martin-Blanco ) . Larval tracheae were dissected at either L2 or L3 larval instar and immunostained according to standard protocols . The following primary antibodies were used: mouse anti-Hdc ( 1:1 ) , mouse anti-β-galactosidase 40 . 1a ( 1:200 ) from the Hybridoma Bank , rabbit anti-PH3 ( Ser ) ( 1:100 ) from Cell signaling , rabbit anti pMad ( 1:100 ) a gift from Gines Morata and goat anti-GFP ( 1:500 ) from Abcam . Secondary antibodies labeled with Alexa 488 , Alexa 555 , or Alexa683 were obtained from Molecular Probes . Micrographs were acquired with Leica SP5 and SPE confocal microscopes and images were processed with FIJI . In order to assess transcriptional activation of snoo , we utilized the SH1402 lacW insertion in the snoo locus . Tracheae from L2 and L3 larvae were assayed for LacZ expression . In order to knockdown Snoo , virgin UAS-snooRNAi flies were crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . In order to overexpress Snoo gene throughout the trachea , virgin UAS-snoo flies were crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . The resulting progeny were reared at the permissive temperature for GAL80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for GAL80 ( 29°C ) and maintained until late L3 and assayed for cell division and Hdc expression . In or der to coexpress Snoo and Fzr , virgin Uas-snoo;Uas-fzr flies were crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and marker expression . In order to assess the accumulation of Snoo protein in the presence of Fzr , virgin UAS-snoo::GFP flies were crossed with btl-Gal4;tub-Gal80 males . In order to continuously express Snoo::GFP larvae were reared at the non-permissive temperature for Gal80 ( 29°C ) until late L3 and then assayed for GFP expression . For the pulsed expression of Snoo::GFP , larvae were reared at the permissive temperature for Gal-80 ( 18°C ) until late L2 , shifted to the non-permissive temperature for Gal80 ( 29°C ) for one day and then shifted back down to the permissive temperature for Gal80 ( 18°C ) until late L3 and assayed for GFP expression . In order to observe the effect of coexpression of Snoo::GFP with Fzr , UAS-snoo::GFP flies were crossed with UAS-fzr flies . The resulting UAS-Snoo;UAS-fzr males were crossed with virgin btl-Gal4;tub-Gal80 flies . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and GFP expression . In order to assess the regulation of stg via Snoo , UAS-snoo flies were crossed with Stg::GFP flies and UAS-snooRNAi flies were crossed with stg-lacZ flies . UAS-snoo;Stg::GFP males were crossed with virgin btl-Gal4;tub-Gal80 flies . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and GFP expression . stg-lacZ; UAS-snooRNAi males were crossed with virgin btl-Gal4<UAS-gfp;tub-Gal80 flies . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and LacZ expression . In order to assess activation of the Dpp pathway in DAP cells as well as expression of Dpp pathway components , tracheae from Dad::n::GFP , dpp-Gal4<UAS-GFP , Tkv::GFP and Mad::GFP were taken from L2 and various stages of L3 larvae and assayed for GFP expression . Wild-type tracheae were also stained for phosphorylated Mad ( pMad ) during L3 . In order to assess the role of Dpp signalling in DAP cell proliferation and progenitor behaviour , virgin UAS-tkvRNAi and UAS-madRNAi flies were crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . The resulting progeny were reared at the permissive temperature for GAL80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for GAL80 ( 29°C ) and maintained until late L3 and assayed for cell division and Hdc expression . In order to assess the regulation of stg via Tkv , UAS-tkvRNAi flies were crossed with stg-lacZ flies . stg-lacZ;UAS-tkvRNAi males were crossed with virgin btl-Gal4<UAS-gfp;tub-Gal80 flies . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and LacZ expression . In order to precociously activate the Dpp pathway in L2 larvae , virgin UAS-tkvQ253D flies were crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . The resulting progeny were reared at the permissive temperature for GAL80 ( 18°C ) until L1 , when they were shifted to the non-permissive temperature for GAL80 ( 29°C ) and maintained until late molting L2 , early L3 and late L3and assayed for cell division and Hdc expression . In order to assess dpp pathway activation in the absence of the Prothoracic Gland , a GAL80 suppressible phantom-gal-4 ( P0206-GAL-4 ) was used to drive Reaper expression in the Prothoracic Gland during the 2nd larval instar . Virgin UAS-rpr;tub-gal80 flies were mated with P0206-Gal4 males . Larvae were maintained at the permissive temperature for GAL80 ( 18°C ) until the 2nd larval instar when they were shifted to the non-permissive temperature ( 29°C ) and maintained there for at least two days at L3 before being assayed for phosphorylated Mad . In order to assess the effect of prolonged JH signaling on Dpp pathway activity as well as DAP cell proliferation and progenitor features , Dad::n::GFP larvae were treated with methoprene [25] ( 21 . 1ng/ul ) applied topically once daily starting at L2 until late L3and assayed for GFP expression , cell division and Hdc expression . In order to test the necessity and sufficiency of Br proteins for the expression of Hdc , virgin UAS-brRNAi , UAS-brZ1 , UAS-brZ2 , UAS-brZ3 . UAS-brZ4 flies were crossed with crossed with btl-Gal4<UAS-gfp;tub-Gal80 males . The resulting progeny were reared at the permissive temperature for GAL80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for GAL80 ( 29°C ) and maintained until late L3 and assayed for cell division and Hdc expression . In order to test the regulation of br-Z2 by Mad and Snoo , Br-Z2::GFP flies were crossed with UAS-snooRNAi and UAS-madRNAi flies . The resulting UAS-snooRNAi/Br-Z2::GFP and UAS-madRNAi;Br-Z2::GFP males were crossed with virgin btl-Gal4;tub-Gal80 flies . The resulting progeny were reared at the permissive temperature for Gal80 ( 18°C ) until L2 , when they were shifted to the non-permissive temperature for Gal80 ( 29°C ) and maintained until late L3 and assayed for cell division and GFP expression . The matScan software [26] was used to scan promoter regions of 2000 nucleotides upstream of the genes of interest for possible binding sites for known transcription factors . Results were filtered by the default value of the matScan software ( hits above 80% of the maximum possible score value ) . Hits were also filtered according to mean conservation scores , requiring a minimum score of 0 . 9 . To further refine the candidate hits , we performed a permutation test by randomly selecting 500 regions of 2000 nucleotides from the dm3 genome and repeating the analysis for all PWMs . We used a cutoff of 0 . 05 for the resulting p-values . Genomic coordinates were obtained by the biomaRt package of Bioconductor [33] with the dm3 version of the Drosophila melanogaster genome . We used 652 position weight matrices ( PWM ) from the JASPAR database [34] downloaded from ( http://pgfe . umassmed . edu/ffs/DownloadPWM . php ? Type=FM&&IsPublic=1 ) . We used the evolutionary conservation track from UCSC [35] for 14 species close to Drosophila melanogaster . | An important feature of organs is their ability to maintain their structure and function in spite of natural or accidental cell loss . This capacity is often sustained by so-called stem cells , which are able to provide new cells of the different types in the organ . In addition , some specialized cells , known as facultative stem cells , also retain the ability to re-enter the cell cycle and replace lost tissue . This process has to be very precisely regulated to provide for the maintenance of the tissues and organs while preventing uncontrolled cellular growth . We have analysed this mechanism in the Drosophila trachea; there , a group of Differentiated Adult Progenitor cells ( or DAP cells ) share the features of facultative stem cells as they remain quiescent during larval growth , reactivate their proliferation at the last larval stage and give rise to the different cell types of the adult tracheal network during metamorphosis . The DAP cells , conversely to the majority of the larval cells , do not enter endocycle and by doing so they acquire the features of adult progenitor cells . In this paper we identify the regulatory mechanism that integrates spatial and temporal cues to precisely activate the tracheal adult progenitor program . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"respiratory",
"system",
"model",
"organisms",
"stem",
"cells",
"membrane",
"receptor",
"signaling",
"hormone",
"receptor",
"signaling",
"epigenetics",
"trachea",
"drosophila",
"research",
"and",
"analysis",
"methods",
"genetic",
"interference",
"animal",
"cells",
"gene",
"expression",
"dpp",
"signaling",
"cascade",
"insects",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"rna",
"anatomy",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"metamorphosis",
"cell",
"signaling",
"larvae",
"organisms",
"signaling",
"cascades"
] | 2016 | Snoo and Dpp Act as Spatial and Temporal Regulators Respectively of Adult Progenitor Cells in the Drosophila Trachea |
We examined recurrent Buruli ulcer cases following treatment and assumed cure in a large cohort of Australian patients living in an endemic area . We report that while the recurrence rate was low ( 2 . 81 cases/year/1000 population ) , it remained similar to the estimated risk of primary infection within the general population of the endemic area ( 0 . 85–4 . 04 cases/year/1 , 000 population ) . The majority of recurrent lesions occurred in different regions of the body and were separated by a median time interval of 44 months . Clinical , treatment and epidemiological factors combined with whole genome sequencing of primary and recurrent isolates suggests that in most recurrent cases a re-infection was more likely as opposed to a relapse of the initial infection . Additionally , all cases occurring more than 12 months after commencement of treatment were likely re-infections . Our study provides important prognostic information for patients and their health care providers concerning the nature and risks associated with recurrent cases of Buruli ulcer in Australia .
Mycobacterium ulcerans ( M . ulcerans ) causes a necrotising infection of skin and soft-tissue known as Buruli ulcer . [1] Since the regular use of antibiotics for Buruli ulcer treatment in Australian populations was introduced at the turn of the century , treatment success rates have been very high . [2–4] Disease cure has assumed to occur if lesions have healed and there have been no recurrent lesions within 12 months of commencing treatment . [1 , 5] However , disease recurrence is known to occur . [6] At present there is no information from the Australian setting on the risk of recurrent disease following treatment and assumed cure , despite this being important prognostic information for patients , their families and health-care providers . Furthermore , it is also not known if recurrent disease represents a late relapse of the initial treated infection or a subsequent re-infection . Clarifying this issue may shed some light on the effectiveness of current treatments if recurrent lesions represent late disease relapse . On the other hand , if they represent re-infection , this may shed some light on the effectiveness of an individual’s immunity against new infections following eradication of an initial M . ulcerans infection , as well as ongoing transmission risk in the community . For the first time , whole genome sequencing has recently been used to examine this issue in four cases of recurrent M . ulcerans disease in Benin , Africa , and suggested that three of the cases represented disease relapse and one re-infection . [6] The aim of our study was to determine the risk of recurrent M . ulcerans lesions following treatment and assumed cure in an Australian population and to use whole genome sequencing techniques combined with clinical , treatment and epidemiological data to determine whether recurrent lesions represented late disease relapse or re-infection .
This study was approved by the Barwon Health Human Research and Ethics Committee . All previously gathered human medical data were analysed in a de-identified fashion .
A total of 426 patients with M . ulcerans were managed at Barwon Health during the study period and included in the analysis . The median age was 57 years ( IQR 37–73 years ) and 225 ( 52 . 8% ) were male . Thirty-four ( 8 . 0% ) patients had diabetes and 35 ( 8 . 2% ) were immune suppressed . Lesions were classified as World Health Organization ( WHO ) category one for 79 . 3% , category two for 10 . 6% and category three for 10 . 1% of lesions . The clinical type of lesion was classified as an ulcer for 85 . 1% , nodule for 6 . 1% , oedematous for 7 . 8% and plaque for 0 . 9% . The median duration of symptoms prior to diagnosis was 42 days ( IQR 28–75 days ) . Of this cohort , seven ( 1 . 6% ) patients were diagnosed with a recurrent lesion ( Table 1 ) . This was over a combined follow-up time since commencement of treatment until the time of study analysis ( 12/4/18 ) of 2813 years , with a median follow-up time of 5 . 7 years ( IQR 3 . 3–9 . 4 years ) . The rate of a recurrent lesion was 2 . 81 per 1000 person years ( 95% CI 1 . 19–5 . 22 per 1000 person years ) ( Fig 1 ) . There were no significant differences in the baseline characteristics between those with a recurrence and those without a recurrence . ( Table 2 ) The recurrent lesions occurred a median 44 months ( IQR 16–68 months ) after treatment commenced for the initial lesion; 5/7 recurrences occurred at least 3 . 4 years from the initial lesion . Four ( 57% ) recurrences were on a completely separate limb and side of the body , one was on the same limb but different region of that limb and 2 were on the same limb and in the same region . Treatment of the initial lesion involved surgery alone for 1 patient , antibiotics alone for 2 patients , and antibiotics combined with surgery for 4 patients ( Table 1 ) . According to the initial treatment , 3/7 ( 43% ) patients were assessed as having a ‘significant risk’ of relapse; patient #1 had only 37 days of combined antibiotics alone , patient #4 had excision combined with antibiotic monotherapy with clarithromycin , and patient #6 had excision alone without adjunctive antibiotics and had positive surgical margins . Whole genome pairwise comparisons of the paired isolates revealed close genetic similarity between pairs ( Fig 2 ) . Indeed , based on our SNP analysis the paired isolates from the patients #3 ( mu77/mu489 ) and #2 ( mu327/mu432 ) were genetically identical ( Fig 2 , Table 3 ) . In contrast , paired isolates from patients #1 , #4 and #5 , contained SNP differences between each pair ( Fig 2 , Table 3 ) . To put this genetic variation in context , we also performed SNP analysis on an additional six unrelated human M . ulcerans isolates from the same endemic area . Three of the six isolates ( mu74 , muJKD8049 , mu08009899 ) were genetically identical to each other following SNP analysis ( Fig 2 ) . Three isolates ( mu146 , mu_UK35 and mu487 ) from the paired cases were also genetically identical to these isolates demonstrating that even apparently unrelated isolates can share a common genotype . Furthermore , this genotype appeared the most dominant within the Bellarine Peninsula isolates we examined . Within two of the three pairs that contained this ‘common’ genotype ( #1 and 5 ) , the primary isolate was more genetically divergent from this ‘common’ genotype compared to the second ( reoccurring ) isolate . The time interval between recurrent lesions did appear to greatly influence the number of SNP differences between the isolates .
Our study has shown that Buruli ulcer has a low recurrence rate in treated Australian patients with an assumed cure living in an endemic region . This provides important prognostic information for patients and their health providers , and may help alleviate the often substantial fears that patients have of becoming reinfected once their initial lesion has been cured . Although the low risk is reassuring , the fact that it can occur means that patients and clinical staff need to be educated and aware of this possibility , so that any recurrent lesions are assessed and diagnosed early when lesions are small , enabling less complex treatment with better outcomes [5] . It is also important to recognise that recurrent lesions can occur many years later and commonly occur on completely different regions of the body compared to the initial lesion . In our study we did not detect an increased risk of recurrent lesions associated with patient characteristics which included age , gender , WHO category and type of lesion , diabetes , immune suppression and the duration of symptoms prior to diagnosis . Although we did not examine host genetics , previous studies have identified genetic factors associated with increased susceptibility to M . ulcerans that may influence the risk of recurrent disease . [14 , 15] We would suggest future studies be performed to assess whether host genetics can predict those at risk of recurrences , or whether this is more likely determined by the intensity of re-exposure . The whole genome sequence analysis revealed a mix of genetic relationships between isolates . Paired isolates from some patients ( #2 and #3 ) were genetically identical , possibly suggesting either late relapse of the initial infection or re-infection from a genetically homogenous source . In the case of patient #3 , the extended time between recurrence ( 46 months ) , the fact that the patient received highly effective treatment , and the fact that the lesions were identified in different body areas ( right forearm and left ankle ) , suggests that re-infection from a genetically homogenous source was more likely . While it’s hard to estimate the degree of genetic change that would occur during a latency period in vivo , we assume that some mutations would occur with longer periods ( particularly 46 months ) . In contrast , the isolates from patient #2 –also genetically identical–were only separated by 12 months , and occurred on the same body region . In this case , a late relapse of the initial infection would appear more likely . There were genetic differences between three of the paired isolates ( patients #1 , #4 , and #5 ) which can be interpreted in two ways . Firstly , it’s possible that they are the result of re-infection from a genetically heterogeneous population . In support of this hypothesis , our previous research examining family clusters of M . ulcerans cases in Australia suggests that exposure risk to M . ulcerans is short-term and may not necessarily be from a genetically homogeneous source [10] . However , given that M . ulcerans is highly clonal in Australia , with only minor genetic variation [13 , 16] , it is expected that some re-infection cases will also be from genetically identical sources . The case of patient #3 , discussed above , would be an example here . The second possible explanation is that the bacterium genetically evolves during its latency period in vivo and thus the cases represent late disease relapse despite a small number of SNP differences . In the case of patients #1 and #5 this latter hypothesis cannot be ruled out , but seems unlikely as in both cases the primary ( first ) isolate had genetically diverged more from the ‘common’ dominant genotype compared to the second isolate . This is further supported in patient #1 by the long duration between lesions ( 44 months ) and in patient #5 by the recurrent lesion being situated on a completely different body area and the initial treatment being highly effective for curing BU . Combined , these findings suggests that re-infection with a different genotype was the most plausible explanation for the #1 and #5 cases . In comparison with the other known study by Eddyani et al . from Africa [6] that looked at recurrent BU cases post treatment between 1989 and 2010 using whole genome sequencing , their recurrence rate ( 100/4951 cases; 2 . 0% ) was similar to ours ( 1 . 6% ) . However this study included recurrent lesions occurring from 6 months following treatment meaning their recurrence rate according to our definition ( ≥ 12 months ) would have been lower . With information from clinical , treatment and epidemiological data supported by whole genome sequencing , 80% of our cases were classified as re-infection whereas 75% of their cases were classified as relapse . In the African study , none of the three cases classified as relapse received effective antibiotics against M . ulcerans , putting them at higher risk of relapse [7] , and in 2 of the three cases the isolates were genetically identical . The third relapse isolate differed by only 1 SNP and occurred on the same body region within a short time interval ( 9 . 5 months ) . In their single case classified as re-infection , the second lesion was on a separate body area and the isolate had a 20 SNP difference compared to the original one . Thus their interpretations were similar to ours whereby the one case we classified as relapse ( #2 ) had a genetically identical isolate on the same region of the body within a short time interval ( 12 months ) , whereas those classified as re-infection had a combination of either being genetically distinct isolates ( #1 , 4 , 5 ) , on separate body areas ( #3 , 4 , 5 ) , having had highly effective treatment ( #3 and 5 ) or having a long time interval between cases ( #1 , 3 , 4 ) . From both studies it is evident that whole genome sequencing can be a useful tool in helping to clarify the likelihood of BU relapse versus re-infection post treatment , as has been the case with tuberculosis [17] . The two recurrent cases who did not have paired isolates available for WGS ( #6 and 7 ) were classified as re-infections based on a combination of separate body areas ( #6 and 7 ) , highly effective treatment ( #7 ) and having a long time interval between cases ( #6 and 7 ) . Additionally , our data suggesting that all recurrent cases which occurred more than 12 months after treatment commenced were classified as re-infections , and the only one occurring after 12 months was classified as disease relapse , would support our previous clinical definitions that treatment failure occurs when a recurrent lesion appears within 12 months of commencing treatment[5] . If , as suggested by our study , most recurrent cases result from re-infection , then at least for a proportion of treated patients acquired protective immunity against the development of recurrent M . ulcerans disease does not develop following an initial infection . Interestingly , the rate of recurrence ( 2 . 81 cases/year/1000 population ) was similar to the estimated risk of infection in the general population of the Bellarine Peninsula ( 0 . 85–4 . 04 cases/year/1 , 000 population ) [18] , suggesting that there may be no significant risk reduction against future infection for previously treated patients . This is in contrast to a study from Uganda in the 1970s which suggested an 88% protective effect over 4 years against recurrent M . ulcerans disease in those with a prior history of the disease . [19] A limitation of our study is that we relied on self-presentation or referral to our health service for diagnosis of recurrent lesions more than 12 months after treatment commenced and therefore there is a risk that some recurrent lesions were not captured in our study . However , as we are the only specialised health service in our region managing M . ulcerans it is likely that any recurrent lesions in patients would have be managed at Barwon Health and therefore we feel the risk of missing recurrent lesions would be small . Additionally , as the incidence of M . ulcerans in the Bellarine Peninsula has fallen in recent years , [20] if this reduction relates to reduced environmental pressure for infection we may have underestimated the risk of recurrent lesions that would occur if the pressure had remained constant . It is also recognised that the number of recurrent cases where isolates had WGS performed was small meaning our results need to be interpreted with some caution . Further research involving WGS of more isolates from recurrent cases should be performed to further validate these findings .
There is a low incidence of recurrent Buruli ulcer in treated Australian patients living in endemic regions and the risk is similar to the estimated risk of primary infection within the general population of the endemic area . The majority of recurrent lesions appear to result from re-infection suggesting that for a proportion of treated patients lifelong immunity against M . ulcerans re-infection does not develop . | Mycobacterium ulcerans ( M . ulcerans ) causes a necrotising infection of skin and soft-tissue known as Buruli ulcer . Since the regular use of antibiotics for Buruli ulcer treatment in Australian populations was introduced at the turn of the century , treatment success rates have been very high . However there is no information from the Australian setting on the risk of recurrent disease following treatment and assumed cure , despite this being important prognostic information for patients , their families and health-care providers . Furthermore , it is also not known if recurrent disease represents late relapse of the initial treated infection or a subsequent new infection . In our study we have shown for the first time in Australian patients living in an endemic area that the incidence of recurrent Buruli ulcer following treatment and healing is low , and that this risk is similar to the estimated risk of primary infection within the general population of the endemic area . Furthermore , we have used clinical , treatment and epidemiological data supported by genomic information of M . ulcerans organisms to determine that the majority of recurrent lesions appear to result from re-infection . This suggests that for a proportion of treated patients’ acquired protective immunity against the development of recurrent M . ulcerans disease does not develop from their initial infection . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
] | [
"antimicrobials",
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"drugs",
"tropical",
"diseases",
"microbiology",
"genome",
"sequencing",
"bacterial",
"diseases",
"signs",
"and",
"symptoms",
"ulcers",
"antibiotics",
"neglected",
"tropical",
"diseases",
"pharmacology",
"molecular",
"biology",
"techniques",
"bacteria",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"buruli",
"ulcer",
"lesions",
"actinobacteria",
"molecular",
"biology",
"mycobacterium",
"ulcerans",
"diagnostic",
"medicine",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"human",
"genetics",
"genetics",
"of",
"disease",
"organisms"
] | 2018 | Low incidence of recurrent Buruli ulcers in treated Australian patients living in an endemic region |
Zika virus was reported in the rainforest city of Iquitos , Peru in 2016 . The potential associations between Zika and fetal neurological disorders were reported extensively in the media regarding neighboring Brazil , and led to great concern about the impact Zika could have on people’s health in Iquitos when it arrived . The aim of this study was to explore the knowledge , attitudes , and preventative practices related to Zika virus and its transmission among women of childbearing age in Iquitos , Peru . Six focus group discussions with 46 women of ages 20–35 from an Iquitos district with confirmed Zika cases were conducted to explore: 1 ) knowledge of Zika transmission , its symptoms , and treatment , 2 ) attitudes regarding Zika , including perceptions of risk for and severity of Zika , and 3 ) preventative practices , including awareness of health promotion activities . Participants were knowledgeable about Zika symptoms and knew it was transmitted by mosquitoes , and about half had heard about the association between Zika and microcephaly , but most lacked knowledge about the associated neurological disorders in adults , its sexual transmission , and ways to prevent infection . They expressed concern for pregnant women exposed to the virus and the impact on the fetus . Participants felt at risk of contracting the Zika virus , yet had not changed preventive practices , possibly in part because their perception of the severity of this disease was low . This study reveals knowledge gaps that could be addressed via health promotion messages that might improve prevention practices to help community members protect themselves from Zika virus during this outbreak .
Zika is a vector-borne disease transmitted by a daytime biting mosquito of the Aedes species , predominantly Aedes aegypti [1] . Common Zika symptoms include rash , mild fever , headache and joint pain [1] . Zika has also been linked to severe neurological disorders , including microcephaly affecting fetuses during pregnancy , contracted through trans-placental transmission , as well as Guillain-Barre syndrome in adults [1] . Since 2015 , outbreaks of Zika have been reported in over 60 countries and territories , with more than 750 , 000 suspected and confirmed cases globally [2] . Zika infections have been reported in eleven of the twenty-five Peruvian departments [3]—both along the coast and in the Amazon basin—and in June of 2016 , the first cases of Zika infection were reported in Iquitos , the capital of the Loreto region in the northern Amazon basin [4] . There have been numerous health campaigns in Iquitos to control the Aedes vector and improve knowledge and preventive practices related to dengue [5] , as all four serotypes have been circulating in Iquitos since 1990 [6] and now , Zika . The health campaigns include emergency fumigation , ongoing larvicide application , and health promotion focused on eliminating unused water containers that may become mosquito oviposition sites [7] . The local regional health authority has led public health campaigns to prevent Zika transmission ( via radio , pamphlet , and billboards ) [5] , but these have not been prolific or remarkably different to existing dengue campaigns , and have not focused on the possible sexual transmission . Condom use is the recommended and most effective method for preventing Zika infection through sexual contact [8] , particularly with recent evidence of the important role of sexual transmission of Zika and viral persistence in bodily fluids [9 , 10] . Studies that reported low knowledge of what to do to prevent Zika , also showed poor practice to do so [11 , 12] . A survey in Brazil conducted with women aged 15–49 during the time of the Zika epidemic found that women with a higher education were more likely to avoid pregnancy due to their knowledge of the association between Zika and microcephaly; higher educated women were also more like to follow preventative measures for Zika infection , such as using protected clothing , applying insecticides , using window screens , and removing standing water breeding sites [13] . A knowledge , attitudes , and practices ( KAP ) survey related to dengue conducted in Iquitos also found that higher education was associated to more knowledge about dengue and preventive practices that did not involve expenditures , but that higher socio-economic status was associated with preventive practices that required money expenditures [5] . Other recent studies on Zika have suggested the need to strengthen campaigns , communication , and health promotion in communities at risk for Zika [11 , 12 , 14] . The objective of this study was to examine qualitatively the knowledge , attitudes , and preventive practices associated with Zika and its transmission routes , amongst women of reproductive age , including those who already may be pregnant , in Iquitos , Peru . We focused on examining what women of reproductive age know , given the specific risks associated with Zika during pregnancy .
Iquitos is the most populated city of the Peruvian Amazon basin ( population of 545 , 000 ) [15] and is surrounded by the Amazon , Nanay , and Itaya Rivers , making it only accessible by air or water [16] . The main forms of employment in Iquitos are small businesses in the extractive industries ( lumber , fishing , oil ) , farming , and tourism [17] . This study took place in the Punchana district—one of the four districts in Iquitos . It was selected because of its history of high Ae . aegypti population densities [6 , 16] , because Punchana has been previously identified as the district in Iquitos where dengue outbreaks start and having the highest dengue seroprevalence rates overall [6] , and because of confirmed presence of Zika virus in this area [3 , 4] There is no publication on this yet , but our dengue research team noted that the spatial and temporal progress of Zika transmission after its detection was very rapid in Iquitos ( AC Morrison , personal communication , June 2018 ) , and was similar to patterns observed after the introduction of a novel dengue serotype into the city [6 , 16] . Since the emergence of Zika in Iquitos , surveillance programs have been put into place to combat the risks associated with infection . There are no cases to date of neurological disease in Peru as a direct cause of Zika , however surveillance remains critical [4 , 18] . This was a qualitative study comprised of six focus group discussions ( FGDs ) conducted in June 2017 . A focus group discussion guide was developed and applied by our research team ( see Appendix A ) , all proficient or native Spanish speakers , and included a local coordinator from Iquitos to facilitate and ensure cultural understanding . Two of the co-authors ( VPS and ACM ) have been conducting dengue research in Iquitos for ~35 years combined , including a dengue knowledge , attitudes , and practices survey and FGDs exploring various aspects of dengue transmission and control in the context of Iquitos [5 , 19 , 20 , 21] . The FGD guide was developed by local experts in our research team and focused on qualitatively exploring themes that are common in knowledge , attitudes , and practices ( KAP ) surveys regarding Zika virus [22 , 23] . The main topics explored in the FGDs were: 1 ) knowledge of Zika transmission , symptoms , and treatment , 2 ) attitudes regarding Zika , including perceived risk for and severity of Zika , and 3 ) common preventative practices ( e . g . , use of repellents or condoms ) , and health messages received in the community regarding Zika . Because dengue and Zika viruses share a common vector and have similar symptoms , participants were also probed to compare characteristics of Zika and dengue fever symptoms and Aedes vector control campaigns in general . The Health Belief Model ( HBM ) was used to inform the focus group discussion guide , specifically the themes surrounding attitudes towards Zika , by focusing on perceived risk and severity; perceived risk represents a person’s perceived susceptibility of experiencing Zika virus , while perceived severity refers to the beliefs a person holds concerning the seriousness of Zika virus [24] . Because of our interest in focusing on women’s knowledge and preventive practices , particularly during childbearing years , and because women are more likely to discuss certain topics , such as family planning , in a single sex setting , only women were recruited for FGDs . Purposive sampling was used to recruit women from the Punchana district . This region was selected as there are affiliated research projects in this district and there is known dengue and Zika transmission . Women of prime childbearing years ( 20–35 ) were approached near their homes by a local field worker who has worked in this district before and invited those who met inclusion criteria of age and gender , and who were available at the time , to participate in the FGDs . Once enough women were recruited for the FGDs , recruitment stopped . The research team covered transportation costs to and from the FGD sites . After reviewing the consent form and obtaining verbal consent from each participant , FGDs were audio recorded and later transcribed . There were also two note takers present at each FGD . After each session , the research team met to discuss and summarize the FGDs and compile the notes into one document . These notes were used alongside the transcripts throughout analysis . A codebook was developed prior to data collection based on themes included in the guide . Once the team had reached saturation—the point where the same topics are emerging in each group [25 , 26]–on the main themes being explored , the codebook was edited to reflect additional identified topics . Two researchers coded the discussions using Dedoose version 7 . 6 . 13 , a web application for managing , analyzing , and presenting qualitative and mixed method research data [27] . Blind coding was used on every other transcription , and then compared , to ensure consistency in code application . Any differences were discussed and addressed before continuing with coding . This study was approved by the Institutional Review Board ( IRB ) of a local non-governmental organization , Asociación Benéfica PRISMA ( CE1425 . 17 ) , and Tulane School of Public Health and Tropical Medicine ( #1040307 ) . Since the FGD with each group of women would only be carried out once and there was no reason to document the participants’ names except if we used a written informed consent , the two IRBs approved the use of verbal consent for participants . Prior to starting the FGDs , all participants gave verbal informed consent to participate and to be audio-recorded . All participants were provided with a copy of the consent document , which included contact information for the IRBs and the PI of the project . All members of the research team completed human subjects training .
All participants from all FGDs knew that infected mosquitos transmit Zika virus , but only one or two participants from each focus group knew that the dengue and Zika virus vector , Aedes aegypti , bites during the day . In fact , almost all respondents described the mosquito that transmits Zika to “sleep during the day” and “give problems at night . ” Very few knew about the sexual transmission route: a few participants in three of the six FGDs mentioned this transmission mechanism . In one FGD , a woman suggested that Zika was transmissible through “sexual relations , ” and the other women laughed and shook their heads in disagreement . Almost all focus group participants correctly and consistently named the most common symptoms , including rash , mild fever , headache , joint and muscle pain , and red eyes . In all of the FGDs , participants distinguished Zika by its distinctive rash or “welts” made up of individual bumps grouped tightly together , which spreads all over the body . The rash was noted as the most common way to differentiate Zika from other similarly presenting illnesses in every FGD . These symptoms were frequently compared to those of dengue fever without prompting from the facilitators . One participant described the differences: “it was similar to dengue , but it was different , because it covers you in welts , it made you itchy and you felt a fever and the children could not sleep . That was the difference . ” When asked about duration of symptoms , responses varied , ranging from 3 to 8 days . When participants were asked to share their understanding of treatment for Zika , women mentioned that there “was nothing” that could be done and it was something “one has to fight with refreshments [keeping hydrated] . ” Despite this , two or three participants in each FGD mentioned “paracetamol and chlorphenamine” ( anti-histamine ) as a common remedy for Zika infection . The refreshments refer to juices from local fruits including malba ( a local plant ) , lemon , and coconut , primarily used for keeping hydrated and reducing fever . Many participants in three of the six focus groups also discussed covering body surfaces with malba leaves to lower body temperature . Healthcare seeking was rarely reported . One participant went to the health post and was diagnosed with Zika , based on clinical signs , without a blood test . Another pregnant participant described regular blood draws to check for Zika as a part of her prenatal care . However , the majority of women felt that symptoms were primarily controllable by using home remedies and paracetamol ( available without prescription ) as listed above , and hence , seeking healthcare was not common . About half of the participants in each FGD knew someone who had experienced Zika virus infection; those who knew someone reported knowing this based on symptom recognition , most commonly the presence of the distinguishable rash . In Iquitos , some laboratory testing was done by the Instituto Nacional de Salud ( National Institute of Health ) , but diagnosis was also provided at local health centers without a laboratory diagnosis . A women voiced that there had been “a lot of Zika” and they [she and fellow Iquitos residents] felt they were at risk of Zika , but “the truth is that here in this area , as it is a jungle area , we are already accustomed to it [Zika]” because “there will always be that little animal , the mosquito . ” And that they had learned not to be afraid of Zika , “If we are afraid we will be self-conscious in the house: I cannot go out , I cannot go for a walk because Zika is everywhere . ” The women explained that pressures of mosquito-borne illnesses , such as “going on a walk and suddenly feeling with discomfort that we can contract [Zika] , ” was not a lifestyle they wanted to face . The women appeared to acknowledge the risk of Zika infection , but chose not to live in fear of something they could not control . The fatalistic attitude expressed by the participants was also evident in their head nodding and strong agreement with each other , as noted by the notetakers . Participants expressed more concern for vulnerable populations , such as children and pregnant women , at risk for contracting Zika virus . Women feared more for children than other family members due to their perceived weaker immune systems and higher susceptibility to high fevers and other symptoms of Zika . When prompted to discuss gender differences relating to Zika , two groups suggested that women were more at risk for mosquito bites than men because their “defenses are a little lower” and they spend more time at home , while men are generally in air conditioned rooms at work or in more rural areas with a lower presence of the Aedes mosquito . Although none of the participants reported knowing anyone with fetal complications from an infected pregnant mother , in five out of six FGDs , participants expressed concern for pregnant women exposed to Zika because of potential effects on fetal development including “small heads” and “missing limbs . ” One pregnant woman said , “It mainly affects the babies and the pregnant mothers… . let’s say , if they get Zika , it can be risky for your baby . That is , it can be born with abnormalities . ” There was awareness in five of the six groups of risks for developing fetuses , but no one mentioned neurological disorders associated with adults , such as Guillain-Barre syndrome . When asked about the severity of Zika compared to other febrile illnesses , including malaria , dengue , and chikungunya , all focus groups but one ranked Zika the least severe of all these febrile illnesses . Participants also stated that “malaria is the worst and can kill you” and “if you get malaria , it is worse than dengue and Zika . ” The severity of Zika seemed to be ranked lowest on the spectrum of mosquito transmitted diseases in the community . The most common responses , stated in all six FGDs , regarding ways to prevent transmission of Zika virus were to use mosquito bed nets and to maintain general household cleanliness . Other reported methods used for mosquito prevention included use of mosquito coils and/or repellent , and emptying water containers to reduce breeding sites . One pregnant woman relayed the advice she had received at a talk at her local health post on preventing transmission of Zika , which included “not to have [sexual] contact without protection , to take care of ourselves , to sleep with a mosquito net , to have a clean house . ” When prompted , nearly all FGD participants reported that they had not changed their individual or household preventative practices in response to the Zika outbreak . A few women in each FGD explained that preventative practices were meant for general mosquito elimination , not to target a specific vector borne disease . Prevention of sexual transmission of Zika was only discussed if participants brought up the sexual transmission route . Within the three focus groups that named the sexual transmission route , only two women knew about the recommended waiting period of six months before initiating sexual contact with a Zika infected partner [8] . When prompted , most women reported condom use challenges due to their partners’ preferences: “they protest , but in the end they do it . ” Women said that if they did use condoms , it was primarily for family planning purposes , not for prevention of sexually transmitted infections . One of the two pregnant participants reported avoiding unprotected sex during pregnancy , while the other pregnant participant was not taking any additional precautions to prevent viral sexual transmission during pregnancy . Within our FGD participants , the preferred family planning method was the injection , formally known as the Depo-Provera birth control injection . When asked about family planning , women reported that the conversations with their partners were “not easy . ” Almost all women strongly reported that they did not want more children and that often children were unplanned . Another Zika prevention method mentioned in every FGD was the use of insecticide , in the form of emergency fumigation campaigns , administered variably from year to year . These campaigns are run by the Dirección Regional de Salud ( DIRESA , the regional Ministry of Health ) , who usually fumigate in the early morning hours when the family is typically at home . The main sentiments FGD participants expressed about the household fumigations was the poor efficacy of the insecticide , or as one said , “they might be spraying water , because instead of disappearing , the number of mosquitos are growing , ” and the lack of engagement from the healthcare workers , who often didn’t have conversations with the families . Participants reported that health education is not routinely provided alongside fumigation campaign activities . As one participant described “They don’t know how to share information . They speak quickly and leave . ” Another participant described her desire , and the missed opportunity , to learn more: This desire was echoed by a number of participants who “would like to know more” about Zika prevention . Despite this , the consensus among most participants was that they felt reasonably well informed about Zika , and reported that they primarily relied on television , radio , and verbal and written information from health posts to inform themselves on this topic . Participants in all six focus groups had a very positive attitude towards health messaging in general and they described their media preferences for health information enthusiastically: listening to the radio or television while cleaning , and reading flyers at the health posts during appointments . When asked about who in the family tends to be informed on the topic of health , participants reported that “the women are more conscious because they have more responsibilities” for the health of their family and desired to know as much information as possible to prevent their household from infirmity .
The FGDs were completed one year after the first confirmed case of Zika was detected in Iquitos , when the epidemic was at its end . While there was considerable concern about the possible microcephaly or neurological impact of Zika among public health officials , there have been no cases of Zika-related microcephaly reported in Peru to date [4 , 18] . In contrast , during the last major dengue outbreak ( 2010–2011 ) , deaths and high hospitalization rates raised much concern among the population of Iquitos and were highly reported in the news and media . Whilst there was some media exposure to Zika , it was relatively scarce , and was not associated with fatalities or the severe birth defects observed in Brazil [28] . We found that the FGD participants were aware of the emergence of Zika within their region , making them feel at risk , but not fearful , of Zika . They may not be fearful because of the general mild symptom presentation of Zika that was observed in Iquitos ( e . g . , more rash , but less fever , than with dengue ) , because of a lack of understanding about the possible neurological effects of the virus , or simply because the communities are accustomed to similar arboviruses . A recent survey on Zika in Honduras revealed a similar lack of knowledge regarding association between Zika and potential neurological outcomes , as well as a low risk perception among the community [12] . Generally , the FGD participants presented little knowledge on the sexual transmission route of Zika and the use of condoms to prevent Zika transmission , despite the evidence of viral presence in semen and vaginal secretions and the ability to sexually transmit the virus [9 , 10 , 29] . That said , there was one participant that stated that she was avoiding unprotected sex during pregnancy for this reason . In contrast , a qualitative study in Brazil found that women were adapting their family planning due to the Zika outbreak , as encouraged by health services [30] . This change in family planning practice may well be due to the first-hand experiences participants in Brazil had with the neurological effects of the Zika virus . Using condoms during sex , however , is easier said than done: respondents acknowledged available condom resources , but commented that their partners ( and men in general ) did not like using condoms and needed to be persuaded . Similarly , in another qualitative study about Zika and reproductive decision-making in Iquitos found couples associate condoms with risk , and are not used within committed relationships [30] . The woman’s responsibility for preventing sexual transmission of Zika and men’s aversion to using condoms was also mentioned in a qualitative study in Brazil examining the effect of the Zika virus on family planning practices [31] . Moreover , in previous studies regarding sexual health in Iquitos , researchers described liberal social norms for sex and marriage [32] , but insufficient sexual education in secondary schools [33] . One programmatic implication from these findings would be to increase the understanding of the importance of condom use and its role in preventing Zika transmission via sex , at least during an outbreak . Moreover , promoting condom education could have other benefits , including reducing transmission of sexually transmitted infections and preventing unwanted pregnancies—but both men and women need to see this value . Condom use would serve various purposes , since most participants commented that they did not want any more pregnancies . Despite being aware of Zika in their community , a great majority of FGD participants reported no changes in their preventative practices . Furthermore , some of the preventative practices routinely used for mosquito control reported in FGDs were not effective for preventing the Aedes aegypti vector as these practices centered around preventing bites from nighttime biting mosquitos; this was consistent with findings from a survey-based study in Iquitos in 2015 that found that only 18 . 6% of participants knew that the Aedes aegypti bite during the day , despite years of dengue health promotion [5] . Despite low levels of knowledge about the vector , FGD participants were very knowledgeable about the symptoms of Zika and were able to distinguish Zika symptoms from those of dengue , mainly based on the type of rash . This was also consistent with dengue study findings that found low knowledge about the dengue vector and its transmission , but higher knowledge of dengue symptoms [5] . Based on suggestions from FGD participants , existing DIRESA teams that implement larviciding and fumigating campaigns could engage with household members and give short educational sessions during their routine house visits carried out at 3 month intervals . Our FGD participants were keen for new health information to prevent sickness in their homes and suggested that larvicide and fumigation teams are well placed to provide health education . There was also apparent fatalism demonstrated by the participants in our study regarding their susceptibility to getting Zika and this may be one of the reasons that participants appeared not to make Zika-specific prevention efforts . Zika was perceived as inevitable and there was a certain acceptance of living in a region with high prevalence of Zika and other arboviruses with no cure , only treatment for symptoms . Moreover , the relatively ‘mild’ symptoms reported by the participants—and the fact that there were no reported cases of babies with microcephaly or adults with neurological complications in the news—may have influenced their perception that Zika is not as serious as other , similar illnesses . Finally , despite the perceived higher severity for the fetus , this did not appear to translate into any suggestion of increased or changed preventative methods for pregnant women . Future health education programs within this region could focus on addressing the fatalism expressed regarding the acquisition of Zika . Other studies which have examined feelings of fatalism suggested that future research and interventions should focus on agency and self-efficacy within specific cultural contexts if they are to address how feelings of fatalism affect care-seeking behaviors [34 , 35] . Another way of empowering communities is through participation in health campaigns . One example would be through peer education , in which people are not objects or recipients of educational projects , but participants in this process , who are able to identify their problems and solutions [36 , 37]; for example , involving volunteer community members in health promotion initiatives and clean up activities . Our findings suggest that women in these communities are receptive to health education and have expressed a desire to acquire more information regarding Zika . Anecdotally , following completion of the groups many women asked if there would be more FGDs in the future , and were enthusiastic to participate , share their knowledge , and learn more . Research in the Iquitos community has found that people are willing to spend time and money—albeit limited—on controlling vectors to prevent diseases: people report sprinkling and/or mopping petroleum or creoline daily on their floors to try to keep mosquitoes away [20] . That said , the main preventive practice for Zika is vector control , and beyond large scale fumigation and larviciding done by the regional health authorities , community members are limited to controlling breeding sites in a rainforest—and for many , this may seem like a futile task—since personal insect repellents and household insecticide sprays are too expensive for most . This article should be viewed in light of some limitations . The focus groups were completed until saturation was reached , but the sample size was small and only completed with women from one district of Iquitos . The response rate was not recorded; recruitment consisted of our local research assistant walking through a selected neighborhood in the district and recruiting women between ages 20–35 who were available to meet for the focus group at specific times . In order to elicit opinions on sexual and reproductive health from all participants , future studies could hold a small educational session following the FGDs and then discuss participants’ opinions on sexual health practices and preventative methods once they had acquired the new knowledge . This would have the added benefit of completing some community education sessions alongside data collection . In conclusion , our findings show that whilst individuals were able to identify some aspects of Zika , such as the symptoms and the mosquito transmission route , as well as the risks for fetal development , the knowledge regarding other aspects of Zika were limited , most notably the sexual transmission and the risk for neurological complications among adults . Perceived severity associated to Zika was consistently low within this study population , which may be partly due to the mild or delayed clinical presentation that was observed in Iquitos . Our findings reveal that the women in this study were especially receptive to health information , and keen to increase their knowledge in order to keep their families healthy . There is an opportunity to improve the knowledge within this at risk population regarding potential neurological outcomes , the sexual transmission pathway , and the most effective preventative methods . Further studies are needed to design and evaluate educational innovations to provide individuals in Zika prevalent regions with quality health information and practices to contribute to the prevention of infection during this epidemic . | Zika virus is an arthropod-borne viral disease that has recently caused epidemics in many Latin American countries and is associated with adverse neurologic outcomes and fetal complications . Although the infection is typically mosquito-transmitted , infection is also possible via sexual transmission . This study uses focus group discussions to explore the knowledge , attitudes , and preventive practices of women of childbearing age at risk for Zika in Iquitos , Peru , where Zika had arrived one year earlier ( in 2016 ) . We found that most women knew Zika is transmitted via mosquitoes , but few knew it could also be transmitted through sex . We also found that women were aware about the link between Zika and newborn microcephaly , but none mentioned the possible neurologic impact on adults . We also found that very few women had changed their behaviors to prevent getting infected with Zika . By identifying gaps in Zika knowledge , attitudes , and prevention methods , we can improve health promotion and ensure the preventive messages are relevant for those most at risk . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"health",
"promotion",
"medicine",
"and",
"health",
"sciences",
"zika",
"fever",
"microcephaly",
"pathology",
"and",
"laboratory",
"medicine",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"pathogens",
"tropical",
"diseases",
"microbiology",
"health",
"care",
"viruses",
"developmental",
"biology",
"women's",
"health",
"rna",
"viruses",
"sexually",
"transmitted",
"diseases",
"pregnancy",
"neglected",
"tropical",
"diseases",
"morphogenesis",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"medical",
"microbiology",
"birth",
"defects",
"dengue",
"fever",
"microbial",
"pathogens",
"congenital",
"disorders",
"flaviviruses",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"health",
"education",
"and",
"awareness",
"organisms",
"zika",
"virus"
] | 2018 | “Zika is everywhere”: A qualitative exploration of knowledge, attitudes and practices towards Zika virus among women of reproductive age in Iquitos, Peru |
Biological systems are subject to inherent stochasticity . Nevertheless , development is remarkably robust , ensuring the consistency of key phenotypic traits such as correct cell numbers in a certain tissue . It is currently unclear which genes modulate phenotypic variability , what their relationship is to core components of developmental gene networks , and what is the developmental basis of variable phenotypes . Here , we start addressing these questions using the robust number of Caenorhabditis elegans epidermal stem cells , known as seam cells , as a readout . We employ genetics , cell lineage tracing , and single molecule imaging to show that mutations in lin-22 , a Hes-related basic helix-loop-helix ( bHLH ) transcription factor , increase seam cell number variability . We show that the increase in phenotypic variability is due to stochastic conversion of normally symmetric cell divisions to asymmetric and vice versa during development , which affect the terminal seam cell number in opposing directions . We demonstrate that LIN-22 acts within the epidermal gene network to antagonise the Wnt signalling pathway . However , lin-22 mutants exhibit cell-to-cell variability in Wnt pathway activation , which correlates with and may drive phenotypic variability . Our study demonstrates the feasibility to study phenotypic trait variance in tractable model organisms using unbiased mutagenesis screens .
It is remarkable how biological systems manage to operate consistently despite facing several types of variation , including the intrinsic stochasticity in every molecular process . This ability of a given system to produce an invariable output in the presence of internal and external perturbations is called robustness [1 , 2] . Developmental processes need to be robust to perturbations to achieve balanced growth and morphogenesis . This includes stem cell number regulation , which protects an organism from tissue hyperplasia , while at the same time facilitates tissue maintenance and repair . Recent advances in gene expression and protein quantification with single-cell resolution have suggested a substantial amount of cell-to-cell molecular heterogeneity in biological systems [3] , raising the question of how robustness is achieved at the phenotypic level . Since Waddington , who first discussed developmental variability [4] , there is a growing interest in understanding phenotypic buffering using both theoretical and experimental approaches [5] . To this end , a key goal is to discover which genes influence phenotypic variance ( Box 1 ) . Although it has previously been shown that disruption of single genes can lead to phenotypic variability , most experimental studies have been targeted to unicellular organisms or tested specific candidates , usually heat-shock proteins [6–9] . For example , the chaperone HSP90 is often thought to play a major buffering role by suppressing phenotypic variability in animals and plants , thereby allowing genetic variation to accumulate in a cryptic form [9 , 10] . However , the developmental mechanisms underlying variability upon Hsp90 impairment are not understood . Furthermore , recent evidence has suggested a more complex picture as perturbations of Hsp90 can also reduce phenotypic variance , indicating a dual role for this chaperone as either a potentiator of variability or buffer [11] . To date , genome-wide mutagenesis screens to identify factors shaping phenotypic variability have not been performed in multicellular animals . Therefore , it remains largely unclear: ( 1 ) what are the genes that modulate developmental trait variance as a response to a specific perturbation , ( 2 ) how these genes fit in developmental gene networks , and ( 3 ) what their specificity is to the phenotypic trait of interest within the context of a whole organism . Here , we address these questions using C . elegans as a model . Developmental patterning in C . elegans is highly stereotypical and these animals are thought to be near-eutelic; that is , there is an almost invariant number of 959 somatic cells present in every adult hermaphrodite [12] . Furthermore , the complete lineage of all cells is known , allowing precise tracing of developmental defects with single-cell resolution [13] . Importantly , C . elegans populations are also isogenic due to their hermaphroditic reproductive mode . This eliminates a key confounder when studying phenotypic variance in a population , which is the presence of standing genetic variation . We particularly focus on seam cells , which are found on both lateral sides of the nematode and contribute to cuticle secretion together with the surrounding hypodermis [14] . The seam cells show stem cell-like properties dividing in a symmetric or asymmetric manner during postembryonic development ( Fig 1A ) . More precisely , animals hatch with 10 embryonically born seam cells per lateral side , and that number increases to 16 after the early second larval stage ( L2 ) due to a symmetric division . The cells also pass through a series of reiterative asymmetric cell divisions during all larval stages , after which one daughter cell differentiates into a neuronal precursor cell or fuses with the syncytial hypodermis , while the ( usually ) posterior daughter cell maintains the stem cell potential ( Fig 1A ) . The postembryonic lineage behaviour of seam cells is not uniform along the anterior-posterior axis . Cell division patterns differ , for example , between the head seam cells ( H0–H2 ) and the seam cells in the mid body ( V1–V6 ) or tail ( T ) , and also within these groups of cells . Seam cell development has been shown to be influenced by a combination of transcription factor activities including GATA factors and the RNT-1/BRO-1 ( Runx1/CBFβ ) module [15–18] , conserved signaling pathways such as the Wnt pathway [19] , and the heterochronic gene pathway that regulates developmental timing in C . elegans [20] . Seam cell number in the widely used laboratory reference strain N2 is consistent with a WT mean of 16 cells per lateral side in the early adult . However , this phenotype is not fully invariant , as it is evident by the low penetrance of animals in the population ( typically around 10% ) showing either more or fewer seam cells ( mostly 17 and 15 cells , respectively ) . To explore mechanisms of developmental robustness , we initiated in this study a forward genetic approach to identify mutants showing a significant increase in the variability of terminal seam cell number , indicative of animal-to-animal variability within the population . We demonstrate that mutations in lin-22 , a Hairy/Enhancer of Split ( Hes ) -related bHLH transcription factor , increase seam cell number variance via stochastic loss and gain of symmetric divisions that occur within single animals and occasionally within the same epidermal lineage . Loss of symmetric divisions at L2 give rise to more neuroblasts at the expense of seam cells , while symmetric divisions towards the seam cell fate at subsequent developmental stages increase the seam cell pool . We show that lin-22 is a core component of the seam cell developmental gene network interacting with the Wnt signaling pathway so that lin-22 null mutants show stochastic Wnt pathway activation that correlates with phenotypic variability . We finally study systemic effects in the nematode and show that gain in variability in seam cell patterning is accompanied by loss of stochasticity or no change in other developmental contexts .
To study the genetic mechanisms underpinning the consistency of seam cell number among individuals , we set out to isolate mutants showing an increase in seam cell number variance . To this end , we mutagenised a strain harbouring an integrated scm∷GFP transgene ( wIs51 ) that is commonly used as a seam cell marker [15 , 34] , allowed the F1 animals to produce self-progeny and selected F2 individuals showing an “extreme” seam cell number phenotype , as defined by a seam cell count that is either lesser than 15 or greater than 17 cells per lateral side . This extreme phenotype is very rare ( <1% ) among WT animals . Variability is defined in this screen at the level of the population , so we hypothesised that the selected animals would either show in the next generation a variable seam cell number phenotype ( Vsc ) or alternatively an increase ( more seam cells phenotype [Msc] ) or decrease ( less seam cells phenotype [Lsc] ) in terminal seam cell number ( Fig 1B ) . Changes in mean and variance are not mutually exclusive and the relationship between these 2 measures depends on the developmental system of choice [2] . We found that the seam cell number variability increases when the phenotypic mean departs in any direction ( increase or decrease ) from the average of 16 cells per lateral side . Therefore , to be confident about a variance change we decided to focus on variable mutants ( vsc ) in which the phenotypic mean showed only minimal change compared to the WT ( Fig 1C ) . We also concentrated on mutants showing 2-sided errors—that is , at the same time both an increase and a decrease in seam cell number within the isogenic population—in an attempt to dissect the developmental basis of bidirectional variability ( Box 1 ) . One of the recovered mutants that satisfied the above criteria was vsc1 . This mutant showed a statistically significant increase in seam cell number variance without a drastic change in the mean ( WT = 16 . 05 ± 0 . 33 SD versus vsc1 = 16 . 29 ± 1 . 87 SD ) and phenotypic errors on both sides of the mean ( Fig 1D ) . Interestingly , vsc1 , like other recovered vsc mutants , showed higher seam cell number variability compared to that observed upon impairment of the expression of genes often considered as bona fide buffering factors such as Hsp90/daf-21 ( SD = 0 . 93; S1A Fig ) . We therefore sought to identify these variance modulators and aimed at mapping the molecular lesion in vsc1 . A current method to identify causative mutations from mutagenesis experiments in C . elegans relies on “mapping-by-sequencing” after crossing to the polymorphic C . elegans isolate CB4856 [35 , 36] . We selected homozygous F2 recombinants based on phenotypic similarity of their progeny to the mutant parental strain , relying on a metric of phenotypic variability ( SD ) , the percentage of extremes in the population and the percentage of animals showing 16 seam cells ( S1B Fig ) . After pooling together these lines and whole genome sequencing , we identified a region of around 0 . 5–1 Mb on the left arm of chromosome IV that contained exclusively N2 markers and was thus likely to harbour the causative mutation for seam cell number variability ( S1C Fig ) . Within this mapping interval , we found a 3 kb deletion in vsc1 mutants located at the upstream region of the lin-22 gene ( Y54G2A . 1 ) ( Fig 2A ) . lin-22 encodes a bHLH transcription factor that is related to Hes transcriptional repressors [37 , 38] . Upon Sanger sequencing of that region , we found that the deletion extends to the 5′ end , deleting part of the last exon of the previous gene vrp-1 ( Y54G2A . 3 ) , and includes a 1 . 7 kb insertion aligning to sequences of the downstream gene mca-3 ( Y67D8C . 10 ) , collectively comprising the icb38 mutation ( Fig 2A and S1 Text ) . The following lines of evidence suggested that the icb38 mutation is a new allele of lin-22; therefore , we refer to it as lin-22 ( icb38 ) . First , it has been previously shown that a hallmark phenotype in hermaphrodites upon disrupting lin-22 function is an increase in the number of sensory post-deirid ( PDE ) neurons [37 , 39 , 40] . In WT , 1 PDE neuron is found per lateral side at the posterior body and is derived from the anterior V5 daughter following the first L2 division [13] . We found that vsc1 mutants , similar to other lin-22 mutants generated via Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated protein-9 ( CRISPR/Cas9 ) genome editing or recovered from independent mutagenesis experiments also showed a significant increase in PDE number as monitored by an increase in dat-1∷GFP foci [41] ( Fig 2B and S2A Fig ) . This included putative null alleles ( icb49 , icb50 due to premature stop codons ) , another allele with an introduced stop codon in the third exon ( ot267 ) , as well as a mutation ( ot269 ) of a single nucleotide located 4 , 940 bp upstream from the lin-22 ATG ( Fig 2A ) , in a region that is deleted in vsc1 mutants , suggesting that this promoter region may play some regulatory role . Importantly , we found that all lin-22 mutant alleles as well as RNA interference ( RNAi ) treatment targeting the lin-22 gene and not the upstream gene vrp-1 , led to an increase in seam cell number variability ( Fig 2C and S2B–S2F Fig ) . Importantly , seam cell number variability in lin-22 ( icb38 ) mutants was cell marker-independent as we observed a similar phenotype when using a bro-1∷GFP transgene to label the seam cells [17] ( S2G Fig ) . We also used genome editing to engineer a putative null lin-22 mutation in the CB4856 isolate and found a comparable increase in seam cell number variability and PDE number , suggesting that these phenotypes were independent of the N2 genetic background ( S2H and S2I Fig ) . Seam cell number variability was sex-independent because it was also observed in males ( S2J Fig ) , as well as lateral side-independent , as we found no correlation between seam cell counts obtained from one side of the animal to those for the other ( S2K Fig ) . Because the lin-22 ( icb38 ) mutation mapped to the upstream noncoding region of lin-22 , we went on to study aspects of lin-22 promoter regulation and the spatiotemporal pattern of lin-22 expression . To this end , we first constructed reporter fusions by placing either ( 1 ) the full lin-22 promoter ( approximately 5 kb ) , or ( 2 ) the distal to ATG lin-22 promoter ( approximately 3 kb ) that is deleted in lin-22 ( icb38 ) mutants , or ( 3 ) the proximal lin-22 promoter ( approximately 2 . 2 kb ) remaining in lin-22 ( icb38 ) mutants in front of green fluorescent protein ( GFP ) . We found that the full promoter drove lin-22 expression mostly in the seam and hypodermis ( hyp7 ) , but also to a much lesser extent in the intestine ( Fig 3A and 3B ) . We also observed that the distal lin-22 promoter drove GFP expression in seam cells and hypodermis , indicating that the deleted region in lin-22 ( icb38 ) mutants contained some putative seam cell enhancer activity ( Fig 3A and 3B ) . The proximal lin-22 promoter fusion showed rare GFP expression in the seam but more frequent expression in the intestine ( Fig 3A and 3B ) . Interestingly , within the deleted lin-22 promoter region in lin-22 ( icb38 ) mutants we identified 2 conserved regions: conserved region 1 ( CR1 ) and conserved region 2 ( CR2 ) between C . elegans and other related Caenorhabditis nematodes ( S3A Fig ) . We showed that CR1 , which contains at least 2 putative GATA binding sites ( S3A Fig ) , was required to drive GFP expression in the seam ( Fig 3A and 3B ) . CR1 was also partially sufficient to restore expression in seam cells as 65% of the animals showed some GFP expression in the seam ( Fig 3A and 3B , n = 50 ) , out of which 18% showed GFP expression in all seam cells that is fully reminiscent of the expression driven by the 3 kb distal fragment . To study the endogenous lin-22 expression pattern , we used single molecule fluorescent in situ hybridization ( smFISH ) , which allows detection of single mRNAs , thus providing a quantitative account of gene expression [42] . In WT , we found that lin-22 expression in the seam is restricted to H0–H2 and V1–V4 cells ( Fig 3C and 3D and S3B Fig ) . After cell division , we found that daughter cells show initially comparable amounts of lin-22 expression , both after the L2 symmetric and the subsequent asymmetric divisions ( Fig 3C and S3C Fig ) . However , lin-22 expression was maintained specifically in the posterior seam cell–fated daughter cell late after asymmetric divisions ( Fig 3C and S3C Fig ) . In contrast to the WT pattern , lin-22 expression was completely absent in lin-22 ( icb38 ) mutants in the seam at all developmental stages ( Fig 3D and S3B Fig ) . Instead , we found stronger expression in the intestine ( S3D Fig ) , which is consistent with the increased intestinal expression that we also observed with the proximal lin-22 promoter∷GFP fusion . Remarkably , lin-22 expression in the seam was decreased in the lin-22 ( ot269 ) mutant background in which a single base substitution disrupts a GATA site within the CR1 region at the distal promoter of lin-22 ( Fig 3E and S3A Fig ) . However , lin-22 seam cell expression was increased in lin-22 ( ot267 ) and the putative null allele lin-22 ( icb49 ) ( Fig 3E ) , indicating that LIN-22 may regulate its own expression via negative feedback . To address whether epidermal GATA factors may be upstream regulators of lin-22 , we quantified lin-22 seam cell expression by smFISH in loss of function mutants of elt-1 and egl-18 [18 , 43] ( Fig 3F and 3G ) and elt-1/egl-18 RNAi-treated animals ( S3E Fig ) . In both cases , the number of lin-22 spots detected was decreased . Taken together , our data suggest that the lin-22 ( icb38 ) mutation leads to loss of lin-22 expression in the epidermis due to the deletion of a GATA site-containing enhancer that is found at the distal end of the lin-22 promoter . The developmental regulation of seam cell number relies on the right balance between epidermal cell proliferation and differentiation . These processes can be monitored by using the scm∷GFP transgene and lineage analysis , which relies on following GFP-positive cells and their patterns of division . When a scm∷GFP positive cell divides , the 2 daughter cells initially express GFP; however , GFP expression persists in the daughter cell that maintains the seam cell fate and disappears in the daughter cell that differentiates . Most V lineages contribute 2 seam cells to the terminal seam cell number , with the exception of V5 that contributes only 1 , as it does not undergo a symmetric division at the early L2 stage ( Fig 1A ) . Previous studies had shown that the increase in PDE number in lin-22 mutant hermaphrodites stems from putative homeotic transformations of V1–V4 to that of V5 [37 , 40] , leading to the prediction that lin-22 mutants were likely to show a decrease in seam cell number as opposed to variability . To decipher the developmental basis of the Vsc phenotype , we performed stage-specific phenotypic analysis using fixed animals carrying both the scm∷GFP and dat-1∷GFP markers . We observed that lin-22 ( icb38 ) eggs hatch with 10 seam cells as normal , indicating that the increase in variability happens postembryonically . Following the first L2 division , we found that at least 2 , but up to 4 , V1–V4 cells stochastically generate a neuroblast in all animals analysed ( n ≥ 50 ) , instead of dividing symmetrically to produce 2 seam cells as in WT ( error 1 in Fig 4A ) . However , at the same time , we observed previously uncharacterised defects , namely animals showing occasional H2 symmetric divisions at the L2 stage ( error 2 ) and V1-V4/H cell symmetric divisions at the third larval ( L3 ) and fourth larval ( L4 ) stage ( errors 3 , 4; Fig 4A ) , as opposed to the WT asymmetric divisions for these lineages at these developmental stages . To verify these initial observations and establish complete seam cell lineages , we performed time-lapse microscopy by imaging scm∷GFP worms while growing from eggs to adults in custom-made microchambers and with 20 minute time resolution [44] . We found frequent V1–V4 asymmetric divisions at the early L2 stage , with the anterior daughter contributing ectopic PDE neurons instead of seam cells ( in 32 out of 56 V1–V4 cell lineages analysed ) ( Fig 4A and 4B ) , although in some instances the anterior daughters gave rise to hybrid lineages contributing both neurons and seam cells ( in 9 out of 56 V1–V4 lineages ) , as previously described [40] . Importantly , we validated the symmetric H2 divisions ( in 5 out of 14 of H2 cell lineages ) at the L2 stage . We also validated stochastic symmetric divisions for V1–V4 ( in 21 out of 56 V1–V4 lineages ) and H1/H2 cells ( in 11 out of 28 H1/H2 cell lineages ) at the L3 and L4 stage ( Fig 4A and 4B and S4B Fig ) . Infrequently , we found single V1–V4 cell lineages repeating the L2 division pattern at the L3 stage ( in 3 out of 56 V1–V4 lineages ) or V5 and V6 symmetric divisions at the L3 and L4 stage ( in 3 out of 28 V5–V6 lineages; S4 Fig ) . We conclude that the developmental basis of the cell number variability in lin-22 mutants is the stochastic loss and gain of symmetric cell divisions . Notably , these developmental errors affect the terminal seam cell number in opposing directions and can co-occur within a single animal ( S4A Fig ) and cell lineage ( Fig 4B ) . The lineage analysis suggested that 2 sides of the seam cell number distribution could be partially separated in time , with ectopic neurogenesis occurring exclusively at the L2 stage and gain of seam cell fate happening mostly at the L3 and L4 stage . Therefore , we predicted that if animals were forced to skip the L2 stage , then seam cell number would decrease in WT due to loss of the L2 symmetric division , whereas gain of seam cell fate at the L3 and L4 stage would antagonise the loss of L2 symmetric divisions in lin-22 mutants . To this end , we knocked down the expression of lin-28 , a factor required for events specific to the L2 stage . In keeping with our prediction , we found that although seam cell numbers decreased in lin-28 RNAi-treated WT animals , this decrease was suppressed in lin-22 ( icb38 ) mutants ( Fig 4C ) . Instead , the number of ectopic PDE neurons , which is exclusively determined at the L2 stage , decreased in lin-28 RNAi-treated both in lin-22 ( icb38 ) animals and the WT , as predicted ( Fig 4D ) . We went on to explore gene expression changes that might be associated with the daughter cell fate symmetry defects as predicted based on lineage analysis . To this end , we studied the expression of key genes of the seam cell gene network by smFISH . We first detected the expression of the GATA transcription factor elt-1 involved in epidermal cell specification . We particularly focused on H2 and V1–V4 cells , which are adjacent cells and yet display contrasting developmental errors at the early L2 stage—H2 often divides symmetrically in the mutant as opposed to asymmetrically , and V1–V4 divide asymmetrically in the mutant rather than symmetrically . Transcripts for elt-1 were detected in WT animals in both the anterior and posterior V1–V4 daughters following the L2 symmetric division and mostly in the posterior daughters of H2 ( Fig 5A and 5B ) . In stark contrast , we found that in lin-22 ( icb38 ) mutants the elt-1 expression pattern in V1–V4 daughter cells became asymmetric with the posterior cells expressing more elt-1 than the anterior ( Fig 5A and 5B ) . At the same time , elt-1 expression in H2 was found to convert to a symmetric pattern in lin-22 mutants ( Fig 5A and 5B ) , with anterior daughter cells in the mutant showing an increase in expression compared to the WT . Another candidate we studied is the posterior Hox gene mab-5 , previously shown to expand qualitatively to the anterior side in lin-22 mutants [37 , 45] . In WT , we detected expression at the posterior end of the animal and specific localisation to the posterior V5 daughter cell after the first L2 division ( Fig 5C ) . In lin-22 ( icb38 ) mutants , mab-5 signal expanded anteriorly but mRNA spots were detected in an asymmetric way in posterior V cell daughters and at comparable levels to expression in V5 ( Fig 5C and S5A Fig ) . Therefore , with regard to both elt-1 and mab-5 expression , the V1–V4 gene expression pattern in lin-22 mutants is reminiscent of that in V5 . In both cases , the expression pattern change was very frequent ( S5B and S5C Fig ) , yet variable among cells of a single animal , with some seam cells showing a change in pattern when adjacent cells did not ( S5D Fig ) . We further studied the expression of the GATA factor egl-18 , which acts downstream of the Wnt pathway influencing seam cell fate [18] . In WT animals , egl-18 is enriched in the posterior daughter cell following the L2 asymmetric division of H2 ( Fig 5D ) , and also enriched in the posterior daughter cells following the subsequent V1–V4 asymmetric division ( Fig 5E and 5F ) . In lin-22 ( icb38 ) mutants , we quantified an increase in egl-18 expression in anterior H2 . p daughter cells at the L2 division with 36% ( 8 out of 22 H2 . pa cells ) of egl-18 expression values in lin-22 ( icb38 ) mutants outside the WT range ( Fig 5D ) . Interestingly , this frequency of egl-18 increase in expression approximately matches the frequency of cell fate symmetry observed in lineaging . At the subsequent asymmetric division , egl-18 expression in the mutant also expanded with typically the most anterior of the 4 V cell daughters ( V . paa ) showing higher expression in the mutant as compared to the WT , with 16 out of 68 egl-18 expression values in the mutant V . paa cells being outside the WT range ( Fig 5E and 5F ) . Importantly , although this increase in V . paa egl-18 expression was observed in the majority of the animals analysed ( 19 out of 20 animals ) , in most cases only some V cells were displaying this pattern ( 17 out of 19 animals ) , indicating again substantial cell-to-cell variability ( S5D and S5E Fig ) . Anterior daughter cell differentiation culminates in cell fusion to the adjacent hypodermal syncytium , a process that is mediated by the transmembrane fusogen protein EFF-1 [17 , 46] . In WT , we detected very tightly regulated eff-1 induction in bursts localised to the anterior differentiating daughter cells ( Fig 5G ) . Consistent with the stochastic loss of differentiation in lin-22 ( icb38 ) mutants and the increase in egl-18 expression in the most anterior V cell daughters , we found frequent absence of eff-1 expression in V . paa cells ( 19 out of 35 cells; Fig 5G ) and enhanced fusion defects in eff-1 ( hy21 ) ; lin-22 ( icb38 ) double mutants ( Fig 5H ) . Taken together , these gene expression results highlight the loss and gain of molecular symmetry at different cell lineages and cell-to-cell variability in gene expression . To understand further the mechanistic basis of seam cell number variability in lin-22 mutants and the unexpected increase in seam cell number in particular , we performed gene expression analysis via RNA sequencing . Although we used whole animal tissue in these experiments , we hypothesised that we might be able to pin down some specific changes relevant to the epidermis due to the largely tissue-specific lin-22 expression pattern . Validating the approach , cell fate transformations in lin-22 mutants have been attributed to transcriptional de-repression of neuronal regulators [37] , and we found that key neuronal development genes , such as the bHLH factors lin-32 and hlh-14 , were significantly upregulated in lin-22 mutants ( S1 Data and S6A Fig ) . Consistent with our smFISH results , we found a decrease in lin-22 expression specifically in lin-22 mutants that harbour mutations in upstream regulatory sequences ( S6B Fig ) . We then focused on putative components of the seam cell gene network to identify gene expression changes that may have a direct influence on seam cell patterning ( S6B Fig ) . Interestingly , we found changes in Wnt-related components in lin-22 mutants , which led us to explore further a possible link between lin-22 and Wnt signalling . To address this possibility , we first compared POP-1 localisation between lin-22 mutants and the WT . POP-1 , which is the T-cell factor/lymphoid enhancer factor ( TCF/LEF ) homolog in C . elegans [47] , controls multiple asymmetric divisions during embryo and larval development [48 , 49] . POP-1 has been shown to asymmetrically localise in seam cell divisions with anterior nuclei containing more POP-1 than SYS-1 , leading to unbound POP-1 repressing Wnt-regulated genes , while posterior signalled cells exhibit less localisation to the nucleus , so that all POP-1 is bound to SYS-1 complex , which activates target genes [19 , 50] . We imaged POP-1:GFP localisation at the 4 V cell stage in L2 asymmetric division . For each WT cell pair , we found pronounced nuclear localisation at the anterior daughter nucleus ( a ) ( Fig 6A ) , with only 6 . 6% of pairs showing aberrant localisation such as equal GFP intensity between the 2 daughter cells ( 5 . 3% ) or enrichment at the posterior nucleus ( p ) ( a < p in 1 . 3%; n = 76 ) . However , we found that 30 . 1% of daughter V pairs in lin-22 ( icb38 ) mutants showed aberrant POP-1:GFP polarity ( a = p in 21 . 9% and a < p in 8 . 2%; n = 146 and Fisher’s test P value < 0 . 0001; Fig 6A ) . To monitor Wnt pathway activity directly , we introduced the POP-1 and HMG-helper optimal promoter ( POPHHOP ) marker in lin-22 ( icb38 ) mutants [51] . This marker reports POP-1 binding to a synthetic enhancer and is strongly expressed around the tail and mildly in the most posterior seam cells [51] . Interestingly , lin-22 ( icb38 ) mutants displayed an expansion of Wnt pathway activity with anterior to somatic gonad seam cells , including H cells , frequently expressing the marker ( 10 out of 21 animals in the mutant as opposed to 2 out of 30 in the WT , Fisher’s test , P value < 0 . 01 ) ( Fig 6B ) . Interestingly , the expansion in the POPHHOP expression domain was sporadic and not observed in a graded manner from the highly expressing cells in the tail to the head of the animal . This indicates cell-to-cell variability in Wnt pathway activation along the body ( Fig 6B ) . To address whether this variability in Wnt pathway activity may have any phenotypic consequences for the seam , we sought to establish a correlation between POPHHOP marker activation and seam cell fate . Due to technical limitations , we were unable to follow cells expressing the marker using time-lapse microscopy . However , we focused on the activation of the marker in H cells and a partially penetrant yet distinctive phenotype in lin-22 mutants , which is the presence of supernumerary seam cells in the head region in around 40% of the animals , likely due to symmetric divisions increasing the seam cell pool in that area ( S5F Fig ) . We asked whether animals that show ectopic POPHHOP expression in H cells at the L2 stage are more likely to develop head seam cell clusters . We found that lin-22 ( icb38 ) animals selected for POPHHOP marker activation in the head are more likely to show this phenotype compared to animals not expressing the marker or animals selected at random ( S5G Fig ) , thus variable Wnt pathway activation in lin-22 seam cells may be directly linked to changes in cell fate . To explore further the changes in Wnt pathway activity , we compared the expression of Wnt ligands and receptors between lin-22 ( icb38 ) and WT animals . We found evidence that the Wnt receptor lin-17 , which is normally expressed only at the posterior end of the animal at the L3 stage , is ectopically induced in more anterior H and V seam cells in lin-22 ( icb38 ) mutants ( Fig 6C and 6D and S5H Fig ) . We hypothesised that an expansion in lin-17 expression may lead to more cells receiving Wnt ligands , thus acquiring the seam cell fate . To test this hypothesis , we produced transgenic animals expressing lin-17 under a seam cell promoter and showed that this transgene is sufficient to cause an increase in seam cell number and variance , although the latter is likely to be purely technical due to the unstable nature of the transgene arrays ( Fig 6E ) . Consistent with the decrease of lin-22 expression in egl-18 loss of function mutants , we obtained further evidence that lin-22 also acts downstream of the Wnt pathway , as lin-22 expression was mildly decreased in the double Wnt ligand mutant background cwn-1 ( ok546 ) ; egl-20 ( n585 ) , which is aphenotypic for seam cell number ( Fig 6F ) [52] . Taken together , our data provide support for a novel crosstalk between lin-22 and Wnt signalling . It is possible that regulators identified in our seam cell screen might also increase variation for a number of independent phenotypes , which might be indicative of loss of animal fitness [23 , 33] . To test whether lin-22 mutants show any fitness defect , we quantified brood size in lin-22 ( icb38 ) animals and found no statistically significant difference to the WT N2 ( Fig 7A ) . To assess tissue specificity of the phenotypic variability , we looked into other developmental decisions involving tight control of cell numbers in C . elegans . One case of natural variability in cell numbers in the WT concerns P3 . p , which is the most anterior vulval cell that divides once before fusing with the epidermis in around 50% of the animals . We compared P3 . p division frequency between the WT and lin-22 ( icb38 ) and found that P3 . p division occurs in nearly 100% of lin-22 ( icb38 ) mutants ( Fig 7B ) . We then quantified by differential interference contrast ( DIC ) microscopy the number of Pn . p cells induced to acquire vulval fates and found a very mild increase in lin-22 ( icb38 ) , mainly due to low penetrant P3 . p and P4 . p induction ( S7A Fig ) . We also quantified the number of uterine π cells using lin-11∷GFP as a marker [53] and found no difference between lin-22 ( icb38 ) mutants and the WT ( S7B Fig ) . Last , we quantified the number of intestinal cells using an elt-2∷GFP marker [54] and found a marginal decrease in the number of nuclei in lin-22 ( icb38 ) mutants compared to the WT ( S7C Fig ) . We conclude that lin-22 ( icb38 ) mutants show an increase in phenotypic variability predominantly in the seam .
In this study , we performed a targeted genetic screen in C . elegans to identify factors shaping phenotypic variance . In particular , we screened for mutants showing an increase in epidermal seam cell number variability without a change in the mean . We identified a deletion in the distal promoter of the transcription factor lin-22 as the molecular cause of seam cell number variability . We showed that this deletion removes a seam cell enhancer and , thus , attenuates lin-22 expression to the extent that we could no longer detect any transcript in the seam . Consistent with this finding , the recovered mutant phenocopies other lin-22 null alleles with respect to seam cell number variability . Identifying genes that modulate ( enhance or suppress ) phenotypic variance in a given developmental system is a fundamental problem in biology that has implications for disease and drug discovery [55] . A key question is whether variance modulators are integrated within developmental gene networks or they are superimposed as core cell homeostasis factors influencing variance . There are examples in the literature supporting the latter possibility with the most prominent being the molecular chaperone HSP90 , which is thought to suppress variation for a variety of different phenotypes [9 , 56] . More high-throughput screens in yeast for genes buffering morphological variation have also identified chromatin factors , cell cycle proteins , components of stress response , and essential genes as key components influencing variability [6 , 23] . In a recent example in plants , a mutation in a broadly expressed mitochondrial protease was found to increase variability in sepal size and shape [57] . In this case , organ shape uniformity was shown to arise from spatiotemporal averaging of already variable cellular growth in WT . Our seam cell number variability screen identified a transcription factor , which we placed within the seam cell gene network ( Fig 7C ) . Consistent with our definition of genes modulating variability ( Box 1 ) , lin-22 null mutants show pronounced phenotypic variability in the seam with 2-sided phenotypic errors and no change in the mean . This phenotypic variability is tissue specific , with no evidence for systemic defects; therefore , it is unlikely to be driven by animal sickness . Interestingly , we also demonstrate that the recovered lin-22 mutation and other vsc mutants have more pronounced effects on variability than impairment of Hsp90 . Our ability to isolate vsc mutants is consistent with theoretical work that suggests modulators of variance may be widespread in developmental systems [58] . Related Hes bHLH proteins act in mammals as transcriptional repressors and play a role in the maintenance of stem cells and progenitors in neural and digestive organ development , influencing binary cell fate decisions [38] . They are also relevant to disease as they are thought to maintain the stemness of cancer stem cells [59] . Stochastic variation in the expression of Hes1 in mouse embryonic stem cells influences neuronal versus mesodermal differentiation and contributes to heterogeneous cell responses such as the timing of commitment of pluripotent stem cells to differentiate [60 , 61] . In comparison to canonical HES factors , LIN-22 does not physically interact with the Groucho homologue UNC-37 as it is lacking a Groucho interacting domain [62] . Therefore , it may rely on passive repression mechanisms by competing for binding sites with other bHLH activators . By using a combination of molecular genetics and time-lapse imaging , we studied the underlying developmental basis of phenotypic variability . We confirmed previous observations that lin-22 mutants [37 , 40] , like mutants in related transcription factors in other systems [63] , show extensive ectopic neurogenesis , which in our model correlates with anterior to posterior lineage transformations . More specifically , V1–V4 cells normally undergoing symmetric cell divisions at the early L2 stage in the WT divide asymmetrically in lin-22 mutants similar to V5 , with the anterior daughter cell generating a neuroblast . At the same stage , H2 cells often divide symmetrically in lin-22 mutants in a manner that resembles V1-V4 lineages of WT , giving rise to daughter cells that do not fuse to the hypodermis and retain seam cell potential . These cell lineage transformations are stage specific , because at subsequent developmental stages both H and V cells can divide symmetrically in lin-22 mutants in a pattern that is not seen in WT lineages at this stage . Interestingly , both types of developmental errors occur within a single lin-22 mutant animal and even within the very same cell lineage but at different stages , possibly relying on the availability of other factors that contextualise the lin-22 role . However , these 2 trends show variable expressivity , therefore , which cells generate neuroblasts or show aberrant symmetric divisions varies stochastically in the population . This developmental tug-of-war between loss and gain of symmetric divisions in lin-22 mutants results in lineages losing and/or gaining seam cells , thereby pushing the terminal seam cell number to either side of the population average of 16 cells . In particular , we explored the mechanistic basis of the symmetrisation of divisions in lin-22 mutants , which was a previously unknown phenotype . We showed that lin-22 mutants feature a hyper activation of Wnt pathway in the seam , as is evident from the increased expression of the downstream target egl-18 in anterior daughter cells , which may act to prevent seam cell daughter differentiation to hyp7 [19] . We also found an increase in the expression of the Wnt receptor lin-17 , previously known to modulate asymmetric cell divisions [19 , 52 , 64] , and a Wnt pathway activity reporter . In the Q neuroblast , the lin-17 receptor itself has been shown to be a transcriptional target of the Wnt pathway [65]; therefore , it is possible that lin-17 upregulation in lin-22 ( icb38 ) mutants is either a cause or consequence of Wnt pathway activation . The Wnt pathway acts throughout the nervous system in C . elegans [66] so it is likely that the activation of Wnt also facilitates the ectopic neurogenesis observed in lin-22 mutants . Interestingly , the spatial activation of the Wnt pathway was found to be variable in lin-22 mutants , with some seam cells showing strong expression of the Wnt pathway reporter when adjacent seam cells did not . We propose that this stochasticity in Wnt pathway activation in lin-22 mutants may drive phenotypic variability . We were able to establish a correlation between Wnt pathway activation in head seam cells at the L2 stage and the subsequent development of head seam cell clusters in the lin-22 mutant background . In the future , with the advent of improved markers to visualise Wnt singalling , it will be very interesting to follow cells while they develop and establish direct correlations between Wnt pathway levels and cell fate . It will also be exciting to explore the underlying mechanisms of cell-to-cell variability in Wnt pathway activation in the seam via identifying missing Wnt pathway regulators and dissecting tissue-specific pathway feedback . Previous studies in seam cell development have largely relied on reporter constructs that provide qualitative information . In our study , we have used smFISH for the first time in the seam to demonstrate that lin-22 is specifically expressed in H0–H2 and V1–V4 cells , with a clear boundary between V4 and V5 that is consistent with its anterior developmental role . However , we observed that the full lin-22 promoter fusion drives expression in all seam cells including the posterior V and T cells . This may be due to posttranscriptional regulation absent in the promoter construct; for example , some miRNA regulation as previously described for related bHLH factors in mice [67] or simply due to the multicopy nature of transgenesis in C . elegans . This highlights the importance of studying the endogenous mRNA expression in comparison to transcriptional reporter transgenes . The lin-22 expression pattern also suggests that other developmental factors should act locally in V6 and T cells to inhibit neurogenesis [40] . Furthermore , we found lin-22 expression to be dynamic during seam cell development , showing initially equal expression in both daughter cells post asymmetric division . This is different to the expression of egl-18 and elt-1 , both found to be enriched at the posterior seam-fated cell postdivision [18] . Interestingly , the expression of certain Hes genes oscillates in many cell types and Hes genes regulate the timing of critical biological events such as somite segmentation or the timing of neuronal differentiation [38 , 68] . A key aspect for Hes1 expression oscillation is its rapid degradation and negative feedback from HES1 protein [69 , 70] , with the latter being a feature in lin-22 regulation as well . Such negative autoregulation is thought to provide stability to gene networks [71] and may be important to constrain lin-22 expression variability . To better understand gene expression dynamics and expression pattern periodicity in the seam , it would be intriguing in the future to explore a possible connection with the heterochronic gene pathway [20 , 72] . The expression pattern changes we describe in lin-22 mutants support lineage-specific loss and gain of daughter cell fate symmetry at the molecular level occurring within single animals . A striking example is the H2 . p daughter cells in lin-22 mutants , which shift from an asymmetric towards a symmetric elt-1 and egl-18 expression pattern , while the adjacent V1–V4 cells show the opposite trend . Interestingly , the gene expression changes we describe in lin-22 null mutants are also subject to stochasticity , displaying cell-to-cell variability within a single animal . A key question is to identify how gene expression variability might relate to phenotypic variability [73 , 74] . By comparing the frequency of gene expression defects and the frequency of aberrant cell linages , it appears once more that stochasticity in downstream Wnt signaling is likely to contribute to cell fate changes . For example , the increase in egl-18 expression in H2 . pa daughters is observed at a comparable frequency to the adoption of seam cell fate for these anterior daughters . On the other hand , expression in H2 . pa of the upstream gene elt-1 is increased more frequently than the observed cell fate symmetrisation , which may reflect a higher threshold for downstream Wnt pathway activation . Therefore , the smFISH results together with the Wnt reporter analysis suggest that stochasticity in Wnt pathway activation among seam cells in lin-22 mutant animals may be an important component of the observed phenotypic variability ( Fig 7C ) . Over the last years , C . elegans has been used as a system to quantify the limits of developmental robustness to environmental variation and other perturbations [75–77] . There are several reasons why we decided to pursue genetic screens in the seam . First , different tissues might show different levels of sensitivity to perturbations . Seam cell number is sensitive to stochastic noise; therefore , we reasoned that the increased flexibility of seam-cell patterning would facilitate our efforts to identify mutations increasing trait variance [78] . Another key reason is that phenotyping in the seam is based on fluorescent markers , thus , it is amenable to high-throughput approaches including fluorescence-based animal sorting . It is possible that regulators buffering seam cell number variability act cell autonomously within the seam or influence seam cell behaviour from a distance . Therefore , it will be interesting to explore systemic defects in mutants we recover from our screen . Despite some mild defects in the intestine , lin-22 mutants have a normal brood size and in general there is no other tissue in which we could detect an increase in phenotypic variability as strikingly as in the seam . Consistent with this , seam cell defects in lin-22 mutants were shown to be lateral side-autonomous . Therefore , it is unlikely that variability in the mutant is determined at the organismal level or comes as a side effect of loss of animal fitness . Remarkably , we found that although seam cell number is more variable in lin-22 mutants , P3 . p division frequency becomes less stochastic with almost all animals showing dividing P3 . p cells that do not fuse to hypodermis at the L2 stage . Interestingly , Wnt pathway activation has been previously shown to result in almost 100% P3 . p division frequency [79] . Therefore , consistent hyper activation of the Wnt pathway in P3 . p in lin-22 mutants or higher sensitivity , or some independent posterisation towards a P4 . p fate , which always divides in WT , may explain the effect on P3 . p division frequency . Nevertheless , this highlights that a single mutation may lead to either more variable or more deterministic events in different cell lineages ( Fig 7C ) . It remains a great challenge to dissect all mechanisms of phenotypic variability in multicellular systems and develop a developmental framework towards interpreting such phenotypes . Recent evidence suggests that multiple developmental decisions including stem cell patterning are governed by chance to some degree and buffering mechanisms are needed to operate at the cell or tissue level [80] . We anticipate that by cloning a broad spectrum of mutants derived from our screen and dissecting the underlying mechanisms we will increase our understanding on the genes modulating variance and their relationship to core developmental networks .
The strains used in this study were cultured and handled according to standard protocols [81] . The JR667 strain containing the scm∷GFP transgene ( wIs51 ) is used as a reference on standard NGM plates with OP50 as a food source . A complete list of strains used in this study is presented in S1 Table . EMS mutagenesis was performed according to standard procedures [81] . We screened 30 , 000 haploid genomes to recover seam cell number mutants . Briefly , nematodes were mutagenised in 4 ml total volume of M9 supplemented with 50 mM EMS ( Sigma Aldrich , St . Louis , MO ) with occasional rotation , then washed 10 times and plated for 1 hour to recover . F2 animals with extreme seam cell counts were selected either using a worm sorter at a speed of 10 animals per second ( Union Biometrica , Holliston , MA ) or manually under a stereomicroscope ( Axio Zoom; Zeiss , Oberkochen , Germany ) using CO2 and the following set up to immobilise animals: a 150 ml conical flask contained in a Styrofoam box was filled with 50–100 ml of absolute ethanol , and a 5 mm diameter rubber tube was fitted on a petri dish lid and the other end was connected to the flask . Dry ice was added until the temperature of the ethanol equilibrated and a constant flow of gas CO2 was achieved . Whole genome sequencing was performed using various Illumina platforms at 20- to 30-fold genome coverage and mapping was performed using the Cloudmap pipeline on Galaxy [82] . lin-22 ( icb38 ) was backcrossed 4 times before phenotypic characterization . The ot269 mutation is a C to T change at −4 , 940 from lin-22 ATG ( TTTTATCTTGATTTACGTGT ) . The icb49 and icb50 alleles are deletions of a single ( ATTGAATCCG-TGGTGGAATCTC ) or 5 nucleotides ( ATTGAAT-----GGTGGAATCTC ) within the first exon of lin-22 . The ot267 is a G to A change within the third exon ( TCCAAATGGGAAAAAGCT ) . The icb49 , icb50 alleles lead to early stop codons , so we refer to them as putative null . The icb49 lin-22 allele was also recovered in CB4856 through independent injections of the same sgRNA targeting lin-22 in N2 . For light and fluorescence microscopy , animals were mounted on 3% agar pads in M9 containing 100 μM sodium azide ( NaN3 ) , covered with a coverslip and viewed under an epifluorescence Ti-eclipse ( Nikon , Minato , Tokyo , Japan ) microscope . Seam cell and PDE neuron numbers were scored at the early adult stage using 1 lateral side per animal . Lineaging analysis was performed by synchronising animals containing both the scm∷GFP and dat-1∷GFP markers by egg laying over a period of 1 hour and observing them at different time points using epifluorescence microscopy . Time-lapse microscopy was performed as previously described [44] . For the POPHHOP selection experiment ( S5F Fig ) , lin-22 ( icb38 ) mutants carrying the POPHHOP reporter and a red seam cell marker were synchronised by bleaching . Individual L2 animals were mounted on 5% agarose pads , anesthetized by using 10 mM muscimol and classified based on presence or absence of nuclear GFP signal in H cells , while keeping track of the lateral side by using the rectum as a reference . Animals were then grown individually and scored at the early adult stage for seam cell number and presence of H seam cell clusters . Synchronised nematode populations were produced by bleaching . Animals were fixed at the appropriate stage as directly monitored by microscopy and smFISH was performed as previously described [76] using a pool of 25–48 oligos fluorescently labelled with Quasar 670 ( Biosearch Technologies , Novato , CA ) . Imaging was performed using a motorized epifluoresence Ti-eclipse microscope ( Nikon ) and a DU-934 CCD-17291 camera ( Andor Technology , Belfast , United Kingdom ) acquiring 0 . 8 um step z-stacks . Image analysis and spot quantification were performed on raw data using a MATLAB ( MathWorks , Natick , MA ) routine as previously described [76] . For the images presented in the results section of this study , the probe signal channel was inverted for clarity ( black spots correspond to mRNAs ) and merged to the seam cell or DAPI fluorescence channel using ImageJ ( NIH , Rockville , MD ) . A complete list of smFISH oligo probes is presented in S2 Table . Larvae were synchronized by bleaching and grown to L3 stage ( 31 hr posthatching ) before total RNA was extracted using TRIzol ( Invitrogen , Carlsbad , CA ) reagent . RNA quality was determined using the Agilent RNA ScreenType System on a 2100 Bioanalyzer ( Agilent , Santa Clara , CA ) . The library preparation was done using the TruSeq stranded mRNA library preparation kit ( Illumina , San Diego , CA ) . The sequencing data were processed and aligned to C . elegans reference genome using Bowtie2 [83] . The bam files were used to generate counts using bedtools [84] . The counts were then normalised using DESeq package in R [85] . Differences in gene expression were then calculated using the negative binomial test in the DESeq package ( FDR = 0 . 1 ) . RNA-seq data are deposited in the NCBI GEO under accession GSE101645 . To edit the lin-22 coding region , the co-CRISPR strategy was used [86] . An sgRNA targeting the following sequence ( ACTGAAATTGAATCCGATGG ) in the first exon of lin-22 was cloned into pU6∷unc-119_sgRNA vector by replacing the unc-119 sgRNA as previously described [87] . The injection mix contained peft3∷cas9 at 50 ng/μl , pU6∷dpy-10_sgRNA at 25 ng/μl , pU6∷lin-22_sgRNA at 25 ng/μl , repair oligo template for dpy-10 at 10 pmol/μl , and myo2∷dsRed at 5 ng/μl . F1 animals showing morphological phenotypes indicative of modifications at the dpy-10 locus were examined for presence of multiple PDE neurons . PCR was performed on the F2 animals by using primers lin22-23F/lin22-22R , and the amplified fragment was sequenced to find the nature of the induced mutation . To construct the lin-22 promoter GFP reporters , the following cloning strategy was used . For the full promoter ( plin-22∷gfp ) , a 5199 bp sequence upstream of the lin-22 ATG was amplified by using the oligos lin22-1F and lin22-2Rfusion from fosmid WRM0627dG07 . For the proximal promoter ( plin-22 ( proximal ) ∷gfp ) , a 2 , 180 bp sequence upstream of the lin-22 ATG was amplified by using the oligos lin22-3F and lin22-2Rfusion from the same fosmid . Both amplicons were fused by PCR to GFP∷H2B∷unc-54 3′ UTR amplified previously from a suitable plasmid using oligos GFP-F and unc54-R . Both constructs were injected into N2 animals at 10 ng/ul with myo-2∷dsRed as co-injection marker . For the distal lin-22 promoter ( plin-22 ( distal ) ∷gfp ) , the distal 3040 bp lin-22 promoter , deleted in lin-22 ( icb38 ) , was amplified from the same lin-22 containing fosmid using the primers lin22-17F and lin22-18R carrying restriction sites for StuI and NheI respectively . The amplicon was cloned in the L3135 vector ( Addgene , Cambridge , MA ) as a StuI/NheI fragment creating pDK1 . To create the CR1 deletion ( plin-22 ( distal CR1 del ) ∷gfp ) reporter , pDK1 was used as template to amplify 2 distinct fragments of the distal promoter , excluding the CR1 using primer pairs lin22-17F/ lin-22 _fus_CR1delR and lin-22_fus_CR1delF/lin22-18R . The 2 amplicons were fused by PCR and inserted into L3135 . To create CR1 sufficient ( CR1∷gfp ) GFP reporter , CR1 was amplified from pDK1 by using primers lin-22_CR1REF and lin-22_CR1RER carrying compatible restriction sites and the amplicon was cloned in L3135 . All 3 reporters were injected into N2 animals at 10 ng/μl with myo-2∷dsRed as co-injection marker . To construct a vector to allow seam cell transgene expression , we used the last intron of arf-3 ( arf-3i ) that is contained within the original pMF1 plasmid and is sufficient for seam cell expression . arf-3i and unc-54 were amplified from previously made plasmids using primer pairs arf-3-EcorI/pes-10-R-Fusion and unc-54-F-Fusion/ unc-54-HIII respectively . The 2 amplicons carried a fusion overlap , including a 23 bp sequence tag containing SwaI and PmeI restriction sites , and were fused by PCR . The resulting amplicon was cloned into pUC57 as an EcoRI/HindIII fragment producing pIR5 . To express lin-17 in the seam , lin-17 was amplified from N2 cDNA using primers lin-17a3-p10 F and lin-17u54R , which carried compatible sequences and allowed insertion of the amplicon in a SwaI digested pIR5 via Gibson assembly . The resulting plasmid pDK5[pseam∷lin-17∷unc-54 3’ UTR] was injected at 30 ng/μl with myo-2∷dsRed as co-injection marker . All constructs used in this study were verified by sequencing and at least 2 independent transgenic lines were obtained and compared . A list of all oligos used in this study is presented in S3 Table . Animals were fed with dsRNA expressing bacteria as a food source . Bacteria were grown overnight and then seeded directly onto NGM plates containing 1 μM IPTG , 25 μg/ml ampicillin and 6 . 25 μg/ml tetracycline . To construct a lin-22 RNAi feeding vector , lin-22 was amplified using the TOPO cloning compatible lin22-15F primer and lin22-14R from the fosmid WRM0627dG07 ( Source Bioscience , Nottingham , United Kingdom ) . The amplicon was inserted by TOPO cloning in pDONR/D-TOPO vector , creating the entry vector pENTR lin-22 . The pENTR lin-22 plasmid was used to insert lin-22 in a gateway compatible L4440 vector via an LR reaction . To construct a vrp-1 RNAi feeding vector , vrp-1 was amplified from N2 genomic DNA using primers Y54G2A . 3a F1 and Y54G2A . 3a R1 , first cloned in a pDNR/D-TOPO vector ( Invitrogen , Carlsbad , CA ) and then inserted into a gateway compatible L4440 vector . Both feeding vectors were transformed into Escherichia coli HT115 to be used for nematode feeding . The elt-1 RNAi feeding vector is pAW565 as described in [17] . All other clones used in this study are commercially available from Source Bioscience . | Organisms are exposed to both internal and external perturbations in every molecular process they go through , and robustness—the ability to maintain their systems unchanged—is crucial for their development and survival . However , the processes that keep the variability of cells as low as possible are barely known . The nematode C . elegans is notable for its highly reproducible development , showing an almost invariant pattern of cell division and differentiation during development; it is thus an ideal model organism in which to search for genes that regulate phenotypic consistency among genetically identical individuals . We focus on a group of lateral epidermal cells—the seam cells—which undergo stem cell-like divisions during postembryonic development . These divisions can either be symmetric towards the seam cell fate , acting to increase the total number of cells , or asymmetric , giving rise to one daughter cell that differentiates into its final fate and another one that serves to keep the number of seam cells constant . We show here that mutations in the transcription factor lin-22 increase seam cell number variability due to stochastic conversion of symmetric divisions into asymmetric ones and vice versa during development , thereby altering the number of terminal seam cell number in opposing directions . We also show that the observed phenotypic variability correlates with the stochastic activation of the conserved Wnt signaling pathway . Our work suggests that core components of developmental gene networks modulate phenotypic variability in multicellular animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"rna",
"interference",
"cell",
"division",
"analysis",
"caenorhabditis",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"green",
"fluorescent",
"protein",
"neuroscience",
"animals",
"animal",
"models",
"caenorhabditis",
"elegans",
"luminescent",
"proteins",
"model",
"organisms",
"experimental",
"organism",
"systems",
"bioassays",
"and",
"physiological",
"analysis",
"epigenetics",
"research",
"and",
"analysis",
"methods",
"cell",
"analysis",
"genetic",
"interference",
"proteins",
"animal",
"cells",
"gene",
"expression",
"biochemistry",
"rna",
"cellular",
"neuroscience",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"phenotypes",
"neurons",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2017 | Stochastic loss and gain of symmetric divisions in the C. elegans epidermis perturbs robustness of stem cell number |
Group C orthobunyaviruses are single-stranded RNA viruses found in both South and North America . Until very recently , and despite their status as important vector-borne human pathogens , no Group C whole genome sequences containing all three segments were available in public databases . Here we report a Group C orthobunyavirus , named El Huayo virus , isolated from a pool of Culex portesi mosquitoes captured near Iquitos , Peru . Although initial metagenomic analysis yielded only a handful of reads belonging to the genus Orthobunyavirus , single contig assemblies were generated for L , M , and S segments totaling over 200 , 000 reads ( ~0 . 5% of sample ) . Given the moderately high viremia in hamsters ( >107 plaque-forming units/ml ) and the propensity for Cx . portesi to feed on rodents , it is possible that El Huayo virus is maintained in nature in a Culex portesi/rodent cycle . El Huayo virus was found to be most similar to Peruvian Caraparu virus isolates and constitutes a novel subclade within Group C .
The Orthobunyavirus genus comprises a diverse set of viral species , represented by multiple serogroups , including: Bunyamwera , California , Group C , and Simbu [1] . Their RNA genome includes three segments ( Small [S] , Medium [M] , and Large [L] ) . The L segment encodes a RNA polymerase ( RdRP ) ; the M segment encodes two glycoproteins ( Gc and Gn ) in addition to a non-structural protein ( NS ) ; and the S segment encodes both a nucleocapsid protein ( NP or N protein ) and a non-structural protein ( NSs ) [2 , 3] . Group C viruses were first identified in Brazil around 1950 . Members of the California serogroup , including La Crosse , California encephalitis , Inkoo , and Tahyna viruses , are known to cause disease in humans [4–8] . Similarly , members of the Bunyamwera serogroup , including Cache Valley and Bunyamwera viruses [9 , 10] , Simbu serogroup , including Akabane , Iquitos , and Schmallenberg viruses [11–13] , and Group C , including Caraparu , Itaya , Marituba , and Oriboca viruses [14–16] , are known to cause disease in humans or domestic animals . Because infection with Group C viruses results in a non-differentiated febrile ( dengue-like ) illness and the lack of available diagnostic assays for these viruses , it has been difficult to associate these viruses with human disease . However , a study by Forshey et al . [17] identified 30 cases of human illness associated with Group C orthobunyaviruses , many of them Caraparu-like , and estimated that about 2 . 5% of febrile illnesses in the region were due to infection with an orthobunyavirus . The goal of our study was to sample , sequence and assemble a novel member of the genus Orthobunyavirus that had been isolated from a pool of Culex portesi mosquitoes captured in Peru in order to provide further genomic insights of this potentially disease-causing virus .
The animal work was approved by the USAMRIID Institutional Animal Care and Use Committee . Research was conducted under an IACUC approved protocol in compliance with the Animal Welfare Act , PHS Policy , and other Federal statutes and regulations relating to animals and experiments involving animals . The facility where this research was conducted is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care , International and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals , National Research Council , 2011 . Mosquitoes were captured at Aotus monkey-baited traps as part of an enzootic dengue study conducted in the vicinity of Iquitos , Peru [18] . Mosquitoes were identified to species , pooled ( up to 25 specimens/pool ) , frozen on dry ice , and kept at -70°C until tested for infectious virus . Mosquito pools were triturated in 2 ml of diluent [10% heat-inactivated fetal bovine serum in Medium 199 with Earle's salts , NaHCO3 and penicillin ( 100 U/ml ) , streptomycin ( 100 μg/ml ) , and nystatin ( 100 U/ml ) ] . The suspensions were clarified by centrifugation ( 3 , 000 rpm for 10 min ) and tested for virus by plaque assay on Vero ( African green monkey kidney , ATCC CCL81 ) cell monolayers . A 0 . l-ml aliquot of each original mosquito suspension and a 1:100 dilution of these suspensions were inoculated into duplicate wells of Vero cell monolayers . A second overlay , containing neutral red stain , was added 2 or 6 d later . If plaques were observed , the agar was removed , and the cells washed with fresh diluent and the resulting viral suspensions aliquoted into cryovials and frozen at –70°C . An aliquot of each suspension was inoculated onto confluent monolayers of Vero cells grown in a T-25 culture flask with 5 ml of liquid cell culture medium and observed daily for evidence of cytopathology . Cell cultures showing cytopathic effects were frozen at –70°C . Later , they were thawed , the suspension clarified by centrifugation at 3 , 000 rpm for 5 min , and then stored as 0 . 5-ml aliquots at –70°C for virus identification studies . The Vero passage 2 stock of one of these viruses , PE-M-0139 ( isolated from a pool of 25 Cx . portesi mosquitoes captured in June 2002 ) , was used in these studies . Total RNA from the Vero passage 2-cell culture supernatant was reverse transcribed using random hexamers , and the resulting cDNA was amplified using multiple displacement amplification . A sequencing library was prepared using the Nextera XT protocol , and sequenced on an Illumina HiSeq 2500 instrument . An initial HiSeq run of 47 , 871 , 860 reads was supplemented with a second HiSeq run of 204 , 323 , 558 reads , yielding 252 , 195 , 418 total 100bp paired-end reads ( NCBI BioProject PRJNA290192 ) . Initial analysis of the metagenomic sample involved a de novo assembly and taxonomic classification approach via MetAMOS [19] , IDBA_UD [20] , Kraken [21] and Krona [22] . However , initial inspection of the classified contigs and unassembled reads provided a convoluted picture of sample constituents , with only two reads classified as a member of the genus Orthobunyavirus ( S1 Fig ) . LMAT [23] ( v1 . 2 . 3 ) was run on the dataset , only 5 reads were assigned to the genus Orthobunyavirus . The reads were adapter clipped and quality trimmed using ea-utils , part of MetAMOS [19] ( fastq-mcf command , default parameters ) using the Nextera XT adapter sequence CTGTCTCTTATACACATCT . To complement the de novo approach , putative orthobunyavirus reads were recruited to a diverse set of orthobunyavirus genomes via blastn [24] ( e-value 0 . 1 , word size 7 ) using a custom orthobunyavirus database ( Caraparu , Zungarococha , Oropouche viruses , containing L , M and S segments ) downloaded from RefSeq [25] . The reference-based strategy filtered the nearly 50 million reads down to 234 , 280 paired-end reads ( 0 . 5% of the sample ) ; blast did not report any read alignments to existing S segment sequences . Assembly of the recruited subset was performed with IDBA-UD ( —pre_correction—num_threads 8—step 10 ) ; assembly was also attempted with SOAPdenovo [26] and Velvet-SC [27] , but these produced fragmented assemblies . The assembly was inspected for misassemblies by mapping all recruited reads back to the assembled contigs using Bowtie 2 [28]; a total of 121 , 901 reads mapped to the L segment ( 1762X avg . coverage ) and 29 , 599 reads mapped to the M segment ( 617X avg . coverage ) . Coverage plots of the read mappings were visualized in IGV [29] . One junction in the assembled M segment was found to lack read support and was not consistent with related M segments ( S4 Fig red arrow ) . A second round of recruitment was performed , including reads from the full assembly covering the region containing the erroneous deletion . The misassembled region was corrected after including these additional reads and rerunning IDBA_UD , resulting in consistent read support across both L and M segments . In addition , a full de novo assembly of the 50 million reads was performed ( IDBA-UD [20] , default parameters ) , resulting in 340 , 327 total contigs . Contigs assembled with the full HiSeq dataset were screened against the Human genome ( hg19 ) and Green Monkey ( BioProject PRJNA215854 ) draft sequence to identify host sequence and misassembled contigs . The recruited assembly was compared to the IDBA-UD [20] assembly on the full dataset using NUCmer [30] . Orthobunyavirus ( L and M ) contigs were identified using both blastx and HMMER . An exhaustive search for the 900-1000bp S segment was performed , without success , using HMMER [31] ( HMM profile http://pfam . xfam . org/family/PF00952 , against all contigs using hmmpress and hmmscan ) . Based on known conserved terminal hairpin sequences found in the UTR regions in Orthobunyavirus genomes [33 , 34] , we searched for terminal hairpin sequences ( AGTAGTGTGCT ) near both 5’and 3’ends in the L , M , and S segments ( within the first 300 nt ) using BLAST [24] ( e-value = 10 , word size = 7 ) , to determine if the assembly was complete on both ends . Amino acid ( aa ) sequences were aligned using MUSCLE [35] ( default parameters ) , back translated to the original nucleotide sequences , edited to trim sequences from both ends that could not be reliably aligned , and then realigned with MUSCLE . Phylogenetic trees were subsequently reconstructed for both a global set of 101 orthobunyavirus genomes ( L segment ) and also on six Group C orthobunyavirus genomes ( L and M segments ) , with FastTree2 [36] . Default parameters were used , and bootstrap support was determined by resampling the site likelihoods 1000 times and applying Shimodaira-Hasegawa test [37] . To determine the potential for El Huayo virus to replicate in a vertebrate host , we inoculated three Syrian hamsters intraperitoneally with 0 . 2 ml of a suspension containing 106 . 5 PFU/ml ( 105 . 8 PFU/hamster ) of the Vero passage 2 stock of El Huayo virus . The hamsters were anesthetized daily and three mosquitoes were allowed to take a blood meal from each of the hamsters . These engorged mosquitoes were then triturated individually in 1 ml of diluent and tested for infectious virus by plaque assay on Vero cells as described above . Hamsters were observed for 21 days for signs of illness .
The El Huayo assembly yielded three contigs ( Table 1 ) with alignments to orthobunyavirus sequences , with best hits for all three segments to Peruvian Caraparu strains [38] . We were unable to identify the known terminal hairpin sequences in the UTR regions , suggesting incomplete assembly of the segments and/or increased divergence in the known conserved region . The de novo assembly of the L and M segments with the first HiSeq dataset was more fragmented than the recruitment approach ( 95 contigs vs . 2 contigs ) with >95% of aligned de novo contigs identical to the recruited assembly . However , the recruitment approach significantly reduced depth of coverage ( 50-fold average reduction in coverage for both segments ) , with a more dramatic effect on the M segment ( 100-fold ) compared to L segment ( 5-fold ) , due to the high level of divergence from the reference strain . Differences between the two assemblies were investigated further with dnadiff [39]; the full de novo assembly had multiple small insertions with respect to the reference-recruited assembly . These insertions were found to have high identity hits to Rhesus macaque and Green monkey genomes , yet were lacking from both Caraparu genomes and the reference-recruited assembly . Closer inspection of these insertions identified them as retroviral sequences and contained within likely misassembled contigs ( S3 Fig ) . Phylogenetic analysis of the L segment suggests that this virus is closely related to Caraparu viruses comprising Group C orthobunyaviruses ( Fig 1 ) . We placed it within the Group C phylogeny , consistent with previously published phylogenetic relationships of orthobunyaviruses isolated from Peru [1] . El Huayo virus therefore appears to represent a novel , previously uncharacterized subclade of Group C viruses . Orthobunyaviruses are known to have high rates of reassortment [40] , and although both L and M segments were most closely related to Caraparu virus ( Fig 2 , Table 2 ) , there is increased polymorphism observed in M relative to L compared to other orthobunyavirus genomes . In addition , Caraparu virus strain FMD0783 was found to be the most similar ( nt/aa ) to both the M and S segments , while strain IQD5973 ( from Iquitos , Peru ) was the most similar ( nt/aa ) to segment L . El Huayo virus replicated to moderate titers in Syrian hamsters , with peak viremias of about 107 . 2 PFU/ml occurring on day 3 after infection ( Fig 3 , Table 3 ) . None of the hamsters displayed signs of illness , and all were well 21 days after infection .
This is the first report of El Huayo virus , a novel member of the Group C orthobunyaviruses . Although rarely associated with human disease in nature , Group C viruses are known to cause febrile illness [13 , 41] . The lack of reported cases is almost certainly due to a lack of diagnostic assays available for this group , and members of this group may be responsible for much of the dengue-like illnesses reported in areas of South and Central America where Aedes aegypti are not common [42] . In fact , Forshey et al . [17] estimated that about 2 . 5% of febrile illnesses in the region were due to infection with an orthobunyavirus , but were misdiagnosed as dengue . Culex portesi , the species from which El Huayo virus was isolated , is a common species known to preferentially feed on rodents and marsupials [43 , 44] and numerous viruses , including Caraparu-like viruses have been isolated from this species [45–47] . The ability of El Huayo virus to replicate to fairly high titer in hamsters indicates that like many other Group C virus , rodents may be involved in the natural maintenance cycle for this virus [48] . Thus , the natural cycle for El Huayo virus appears to be between Cx . portesi and rodents in the Amazon Basin region . Because these viruses have a segmented genome , and because genetic reassortment has been demonstrated in this family/genus [49] , the orthobunyaviruses are an ideal model for studying the evolution of novel viruses by genetic reassortment . How reassortment affects disease in humans and the ability of these viruses to replicate in vector species are key open questions . In our initial comparative analysis , the best matches in our reference database shared ~60–80% nucleotide identity and 70–90% identity at the amino acid level with the ( translated ) novel S , M and L segment sequences , respectively . Given the low sequence identity of segment M relative to segment L , segment M might represent a novel reassortment; the region from 1500-2500bp contains a dramatic reduction in similarity to all known segment M strains available in RefSeq . High divergence relative to existing genomes is a challenge for homology detection methods; sensitivity must be increased to detect divergent matches , but the increase in sensitivity also leads to potential misclassifications . Sensitive profile alignment methods based on hidden Markov models can detect protein domain signatures in cases where extreme divergence makes other methods infeasible [18] , such as in the case of the highly divergent S segment recently reported for Brazoran virus [26] which was double the size of previously published orthobunyavirus S segments . Its S segment contained no known homology to existing segment S proteins; however , similar to what we report here , it did have conserved orthobunyavirus domains that were detected via InterProScan [27] . While insufficient sequencing depth in our initial HiSeq run prohibited detection of the S segment , adding another HiSeq run allowed for the detection of this small viral segment . This result highlights that lower abundance sequences in environment samples may often be missed , and sequencing depth is still an important tool for uncovering low abundance novel viruses from metagenomic samples . Based on amino acid sequence similarity , the orthobunyavirus genome of El Huayo virus reported in this study is most closely related to Caraparu virus Peruvian strains IQD5973 and FMD0783 [38] , both recently deposited in GenBank . This recent growth in publicly available Group C orthobunyavirus genomes enabled the reliable placement of our novel strain within the Group C serogroup . Prior to Huang et al . 2014 , there were no complete genomes ( including all three segments ) from within Group C . Lack of complete genomic sequences of serogroups of interest can lead to misclassification or misidentification , evidenced by a recent study that reported that a collection of Group C genomes likely require further validation [38] . This highlights the importance of efforts to populate reference databases . There exists a vast underrepresentation of viral diversity for various clades , and of particular interest to this study , there are only a small number of South American orthobunyavirus sequences . Continuing efforts are required to fill out viral reference databases to ensure reliable identification and characterization of novel Bunyaviridae genomes . An additional confounding factor for novel virus identification and assembly is host endogenous retroviral elements [50] ( S2 Fig and S3 Fig ) . Aggressive assembly strategies can result in chimeric host-plus-virus assemblies in which sequence shared by both virus and host results in false joins between the two genomes; specifically , retroviral elements integrated into host genomes . We have shown that a recruitment-based strategy , even at relatively high levels of amino acid divergence , can prove useful for avoiding co-assembly of host and target virus . However , this approach requires the presence of reference strains in the database and is prone to under-recruitment of reads in highly polymorphic regions . In summary , while advances in sequencing technology allow for the discovery of novel viruses present at low abundances in a sample , care must be taken to properly address confounding factors . | Arthropod-borne viruses remain a significant cause of human and domestic animal disease and new viruses are constantly being discovered . RNA virus discovery and assembly remains a challenge due to highly polymorphic genomes , current lack of breadth and depth of publicly available viral genomes , and confounding factors due to host sequence and sequencing biases . We describe the discovery and genome assembly of El Huayo virus , a group C orthobunyavirus isolated from a pool of Culex portesi mosquitoes captured near Iquitos , Peru , and named for the Jardin Botanicao Arboretum El Huayo near where the Cx . portesi from which the virus was isolated were captured . Although orthobunyaviruses are not commonly associated with human disease , Group C members are widespread in Central and South America and known to cause febrile illness . The discovery , and genome assembly , of El Huayo virus may help to explain numerous dengue-like illnesses where Aedes aegypti are not commonly found . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"invertebrates",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"vertebrates",
"animals",
"sequence",
"assembly",
"tools",
"mammals",
"viruses",
"genomic",
"databases",
"rna",
"viruses",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"mammalian",
"genomics",
"insect",
"vectors",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"hamsters",
"viral",
"genomics",
"sequence",
"alignment",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"biological",
"databases",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"animal",
"genomics",
"arthropoda",
"mosquitoes",
"rodents",
"flaviviruses",
"virology",
"database",
"and",
"informatics",
"methods",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"amniotes",
"computational",
"biology",
"organisms"
] | 2016 | Identification and Genomic Analysis of a Novel Group C Orthobunyavirus Isolated from a Mosquito Captured near Iquitos, Peru |
Related organisms typically rely on orthologous regulatory proteins to respond to a given signal . However , the extent to which ( or even if ) the targets of shared regulatory proteins are maintained across species has remained largely unknown . This question is of particular significance in bacteria due to the widespread effects of horizontal gene transfer . Here , we address this question by investigating the regulons controlled by the DNA-binding PhoP protein , which governs virulence and Mg2+ homeostasis in several bacterial species . We establish that the ancestral PhoP protein directs largely different gene sets in ten analyzed species of the family Enterobacteriaceae , reflecting both regulation of species-specific targets and transcriptional rewiring of shared genes . The two targets directly activated by PhoP in all ten species ( the most distant of which diverged >200 million years ago ) , and coding for the most conserved proteins are the phoPQ operon itself and the lipoprotein-encoding slyB gene , which decreases PhoP protein activity . The Mg2+-responsive PhoP protein dictates expression of Mg2+ transporters and of enzymes that modify Mg2+-binding sites in the cell envelope in most analyzed species . In contrast to the core PhoP regulon , which determines the amount of active PhoP and copes with the low Mg2+ stress , the variable members of the regulon contribute species-specific traits , a property shared with regulons controlled by dissimilar regulatory proteins and responding to different signals .
The ability of an organism to orchestrate responses to environmental changes often depends on transcriptional regulatory proteins that control the expression of multiple genes . Related species typically rely on orthologous regulatory proteins to respond to a given stimulus . However , the extent to which the targets of regulation of such orthologs ( i . e . , the regulon ) are retained across species is not clear . This is of special interest in bacteria due to the rampant effects of horizontal gene transfer ( reviewed in [1] , [2] ) , which raises questions about how bacterial regulons have been shaped by widespread gains and subsequent losses of genes , and about the role played both by the conserved targets of regulation as well as by those that are species-specific . By contrast , in eukaryotes , where most of the experimental studies on the evolution of gene regulation have been carried out , transcriptional rewiring ( i . e . gains and/or losses of interactions between orthologous regulatory proteins and orthologous target genes ) appears to be the main source of variability in regulatory networks [3]–[5] as related eukaryotic species have similar gene content . We have addressed the evolution of bacterial gene regulation by investigating the regulons controlled by the DNA-binding regulatory protein PhoP in several enteric species , the most distant of which shared a common ancestor >200 million years ago . The activity of the DNA-binding regulatory protein PhoP is dictated by its cognate sensor PhoQ , which responds to the extracytoplasmic levels of Mg2+: transcription of PhoP-activated genes is promoted in low Mg2+ and repressed in high Mg2+ [6] . The PhoP regulon has been best characterized in the human pathogen Salmonella enterica serovar Typhimurium , where the PhoP protein regulates ∼3% of the genes [7] both directly by binding to its target promoters , and indirectly by altering the levels and/or activity of other regulatory proteins and systems [8] . These PhoP-activated targets include Mg2+ transporters , enzymes involved in the covalent modification of cell envelope components , as well as virulence proteins whose biochemical activities remain largely undefined . Despite establishing different interactions with their animal and plant hosts , the bubonic plague agent Yersinia pestis , the diarrhea-causing Shigella flexneri and the plant-pathogen Erwinia carotovora depend on a functional PhoP/PhoQ system to cause disease [9]–[14] , like Salmonella [15]–[17] . The phoPQ genes are also found in the human commensal Escherichia coli as well as in the soil dwelling Klebsiella pneumoniae and in Sodalis glossinidius , a secondary symbiont of the tsetse fly . The presence of the PhoP/PhoQ system in this phenotypically diverse group of bacteria suggests that PhoP may regulate the expression of different sets of genes across species ( i . e . , each regulon might be suited to the niche in which each organism proliferates . ) Alternatively or in addition , PhoP could control cellular functions that are shared among different species in spite of their distinct lifestyles . Here we examine the evolution of the PhoP regulon across ten species of the family Enterobacteriaceae . We experimentally determine that PhoP has adopted largely different targets of regulation , the result of horizontal gene transfer events that altered gene content even among closely related species . We show that transcriptional rewiring events involving gains and/or losses of interactions between PhoP and shared genes have shaped the PhoP regulon and may contribute to phenotypic differences between organisms . Moreover , we establish that the core members of the PhoP regulon ( i . e . , those maintained in all analyzed members of the Enterobacteriaceae family ) participate in regulatory loops designed to control the level and activity of the PhoP/PhoQ system; and that those PhoP-regulated target genes common to most species ensure Mg2+ homeostasis . This demonstrates that governing the activity of the PhoP/PhoQ regulatory system is critical for its proper functioning independently of how different its regulated targets might be .
We sought to identify the PhoP-regulated genes in the plague agent Y . pestis because: First , a functional PhoP protein is necessary for virulence in mice and for survival inside mammalian cells [13] , as in Salmonella [15]–[18] . And second , S . enterica and Y . pestis are distantly-related members of the family Enterobacteriaceae , having diverged from their last common ancestor >200 million years ago [19] . To enable the identification of genes regulated by the PhoP protein and to distinguish between genes that are directly and indirectly regulated by PhoP , we used both expression microarray analysis of wild-type vs . phoP mutant strains as well as chromatin immunoprecipitation followed by array hybridization ( ChIP-chip ) on custom-made whole genome tiling arrays ( a Y . pestis strain harboring an epitope-tagged phoP gene in the chromosome was utilized for ChIP and as the wild-type strain in the expression microarrays . ) Three biological replicates of each sample ( RNA or DNA ) were prepared from Y . pestis cells grown in defined medium containing low ( 50 µM ) Mg2+ , which are inducing conditions for the PhoP/PhoQ system , and hybridized to custom-designed NimbleGen tiling microarrays ( Figure S1 ) . Data from the six expression microarrays ( three wild-type and three phoP− ) were combined for downstream data processing whereas the three ChIP-chip data sets were analyzed individually . Based solely on the expression microarray data ( i . e . without including the ChIP-chip results ) , 31 and 14 transcription units appeared to be activated and repressed by the phoP gene , respectively ( Table S2 ) ( see Text S1 for a description of how the tiling microarray data were processed and Table S1 for a list of all the probes that exhibited differential expression ( >2-fold ) ) . We note that a single transcription unit may entail several co-transcribed ORFs ( which is readily inferred from the expression pattern of contiguous probes in the tiling array ) as well as transcripts corresponding to genes not previously annotated . This set of transcripts represents potential direct and indirect targets of the PhoP protein . We considered a transcription unit to be directly regulated by PhoP if there was a ChIP peak <300 bp from the 5′ end of a transcript that appeared phoP-regulated in the expression microarrays . In most cases there were ChIP peaks with significant scores ( FDR≤0 . 05 ) in similar positions in all three biological replicates . We used real time PCR to verify whether these transcripts were indeed phoP-regulated and if the promoter regions of these transcripts were enriched in the ChIP samples ( data not shown ) . Then , we identified their transcription start sites by primer extension or S1 mapping , and determined that these start sites were phoP-dependent . PhoP binding sites were experimentally and/or computationally identified within 100 nt upstream of the transcription start sites . The location and orientation of the PhoP boxes in all these promoters are similar to those in well-characterized PhoP-activated promoters in E . coli and S . enterica [7] , [20] . Sixteen transcripts met all these criteria ( Table S3 ) . Thus , in this study we refer to these sixteen transcripts as the Yersinia PhoP regulon . ( We found ChIP peaks in front of only two repressed transcripts ( Table S2 ) suggesting that PhoP may repress them directly . ) A list of PhoP regulon members in Y . pestis biovar Microtus has been recently reported [21] . Curiously , only four of the 18 promoters described as directly regulated by PhoP overlap with the data set presented here . The reason ( s ) for the discrepancy are not entirely clear , but could be due to the following: First , our approach was based on a genome-wide search for in vivo PhoP binding regions located <300 bp away from phoP-dependent transcripts identified in vivo whereas Li . et al . [21] investigated binding of the purified PhoP protein in vitro to DNA fragments located upstream of ORFs that could be phoP-regulated . Second , Li et al . grew Yersinia under conditions [22] that are different from those shown to induce the Yersinia PhoP/PhoQ system effectively [23] . And third , the Y . pestis strain utilized by Li et al . is not the same as that examined in this work . Importantly , the structures of all the PhoP-regulated promoters identified in our study resemble those of well-characterized PhoP-activated promoters in other organisms [7] , [24] , [25] . By contrast , the only promoters described in [21] that harbor structures reminiscent of previously characterized PhoP-dependent promoters are those overlapping with our data , and the remaining putative PhoP-activated promoters contain PhoP boxes at locations uncharacteristic of a bacterial transcriptional activator ( i . e . , >100 nt upstream of the transcription start site , overlapping with the −10 sequence , or positioned downstream of the transcription start site . ) Based on their occurrence in Y . pestis and S . enterica , we classified the identified PhoP-regulated targets from Yersinia and the 20 transcription units known to be directly activated by PhoP in S . enterica [7] , [24] , [26] into three groups: ( 1 ) genes/operons present in Yersinia but not in Salmonella , or vice versa; ( 2 ) genes/operons present in both species but controlled by PhoP in only one of the two species; and ( 3 ) orthologous genes/operons controlled by PhoP in both Yersinia and Salmonella ( Figure 1 ) . As discussed below , the first group of genes is by far the most abundant and their sporadic distribution in other enteric species suggest that they have been horizontally acquired . The second group of genes ( five out of sixteen ) represent transcriptional rewiring events , a phenomenon that , to our knowledge , has not been experimentally addressed in bacteria . The third group of genes is the smallest set and , as the results presented in the following sections indicate , they appear to play distinct critical roles in the proper functioning of the PhoP/PhoQ system . The majority of genes directly regulated by PhoP in Y . pestis have no BLAST matches in S . enterica and vice versa ( Figure 2 and Table S4 ) . To determine whether the absence of homologs was particular to the Yersinia-Salmonella pairwise comparison , we searched the genomes of eight additional members of the family Enterobacteriaceae for the presence of sequences homologous to the identified PhoP-regulated protein-coding genes of Yersinia and Salmonella . This analysis revealed that over half of the gene products directly controlled by PhoP in Yersinia lack homologs in at least three other species ( absence of homologous sequences is represented by white squares in Figure 2A ) . This finding indicates that these genes have been gained and/or lost during the evolution of the family Enterobacteriaceae . Analysis of the S . enterica PhoP regulon revealed a similar pattern as nearly half of the protein-coding genes directly regulated by PhoP had no homologs in most of the other enteric species ( white area in Figure 2B ) . This is in parallel to the overall differences in gene content that exist between the phylogenetically most distant species in the analyzed group , Escherichia coli and Y . pestis , which share only ∼50% of their genes [19] . Consistent with the conclusions from a purely computational comparison of the E . coli and S . enterica PhoP regulons [27] , our data indicate that orthologous PhoP proteins promote expression of largely distinct gene sets in individual enteric species . While the biochemical function of the species-specific PhoP-activated gene products remains largely unknown , several of the Salmonella-specific targets have been implicated in survival within host cells , such as mig-14 [28] , or in resistance to host antimicrobial products , such as ugtL [29] . The majority of PhoP-activated Yersinia-specific targets encode uncharacterized proteins . We determined the degree of sequence identity for each pair of homologs by calculating their conservation scores ( CS ) [30] , [31] , which represent the BlastP score of the closest homologue in a particular species divided by the BlastP score of the protein against itself . CS values range from 0 when no homolog or ortholog is detected in another species , to 1 when the closest homolog exhibits 100% amino acid identity . This analysis revealed that only three gene products directly controlled by the PhoP protein both in Yersinia and Salmonella – PhoP , PhoQ , and SlyB – are relatively well conserved in the analyzed enteric species ( red squares in Figure 2A and 2B ) . By contrast , the well-conserved psiE , ompX and rstA genes are regulated by PhoP only in one of the two species . A subset of the genes regulated directly by PhoP in either Yersinia or Salmonella encodes products poorly conserved in other enterics ( blue squares in Figures 2A and 2B ) . These genes are unlikely to be true orthologs because the level of amino acid identity between the corresponding products is considerable lower than the median ∼72% identity that is found between E . coli and Y . pestis orthologs [19] ) ( blue in Figure 2 represents roughly 40% amino acid identity ) . For instance , PhoP promotes transcription of the Salmonella pagP gene and the Yersinia y2563 gene by binding to their respective promoter regions ( Figure 2 ) . Even though the pagP and y2563 genes are recovered as the best reciprocal hits in a BLAST search , their gene products are only 42% identical and do not seem to carry out the same biochemical function because the PagP-mediated palmitoylation of the lipid A established in S . enterica [32] has not been detected in Y . pestis [33] . Therefore , even when homologous genes are regulated in a like manner , the activities of their gene products may not be retained across species . The gains and/or losses of interactions between orthologous regulatory proteins and orthologous target genes ( i . e . , transcriptional rewiring ) account for the majority of differences in the targets of regulation among related eukaryotic species [3]–[5] . Rewiring events have not been considered in previous analyses of bacterial regulons because the purely computational comparisons reported to date have focused on gene content , and thus , they assumed that if a regulatory protein controls a particular target in one species , such regulatory relationship will be conserved in another species [34]–[36] . By combining ChIP-chip and expression data obtained with full genome tiling arrays , we could experimentally determine the contribution that rewiring events make to the composition of the PhoP regulons . We established that the PhoP protein controls transcription of both the outer membrane-encoding gene ompX and the regulatory gene rstA in Salmonella but not in Yersinia . This may result in significant phenotypic differences between these two species because the rstA gene encodes a regulatory protein that modulates the levels of the alternative sigma factor RpoS [37] and of a Fur-repressed iron transporter [38] in Salmonella . Conversely , the phosphate-starvation inducible gene psiE , the putative aminidase encoding gene y1877 ( ybjR ) , and the putative inner membrane protein coding gene y2124 ( STM3036 ) are regulated by PhoP in Yersinia but not in Salmonella . Considering that ∼16 transcripts directly regulated by PhoP have homologous sequences in Y . pestis and S . enterica , this means that ∼30% of them have undergone transcriptional rewiring events . To our knowledge , this is the first report on the prevalence of transcriptional rewiring in a bacterial regulon based on experimental evidence . In addition to changes in regulon membership , rewiring events can result in novel interactions between orthologous regulatory proteins and orthologous target genes that are qualitatively similar ( i . e . , a target gene is turned on in response to the same signal ) [39] but generate quantitatively different outputs . We determined that the Yersinia PhoP protein governs transcription of the pbgP ( y1917 ) operon and the ugd ( y2147 ) gene directly , by binding to their respective promoters ( Table S3 , see also [23] ) , but that it does so indirectly in Salmonella ( i . e . , by activating a different regulatory protein ) [40]–[42] . Though the direct and indirect pathways are qualitatively similar in the two species ( i . e . , low Mg2+ promotes transcription of the polymyxin B-resistance conferring pbgP and ugd genes ) , the indirect pathway operating in Salmonella exhibits signal amplification and expression persistence relative to the direct pathway present in Yersinia [43] . The PhoP-activated mgtC gene encodes an inner membrane protein necessary for virulence in mice , survival within macrophages and growth in low Mg2+ in Salmonella [44] , and for the survival of Y . pestis inside macrophages [10] . However , the low level of amino acid identity between the two MgtC proteins , the sporadic phylogenetic distribution of the mgtC gene within the family Enterobacteriaceae [44] as well as the presence of mgtC homologs in more distant bacterial species such as Burkholderia cenocepacia [45] raises the possibility that the Salmonella and Yersinia mgtC genes may be xenologs rather than orthologous ( i . e . , acquired separately by Salmonella and Yersinia in independent horizontal-gene transfer events ) and likely incorporated into the two PhoP regulons independently . This could reflect that the mgtC gene enables survival within mammalian cells and in Mg2+-limiting environments , which are conditions that activate the PhoP/PhoQ system [6] . The Salmonella PhoP/PhoQ system is transcriptionally autoregulated in a positive fashion by the PhoP protein binding to a promoter located immediately upstream of the phoP coding region [46] , [47] . We established that the Yersinia phoP and phoQ genes were turned on in low Mg2+ in a PhoP-dependent fashion ( Figure 3 ) , just like the Salmonella phoP and phoQ genes [6] . However , a PhoP box could not be identified immediately upstream of the phoP coding region in Yersinia and the PhoP protein did not bind to this region in vivo ( Figure 3A and 3B ) . We determined that positive autoregulation of the Yersinia phoP and phoQ genes is mediated by a promoter located upstream of y1795 , the gene located 5′ of phoP and predicted to be transcribed in the same direction as the phoP and phoQ genes ( Figure 3A ) because: First , the PhoP protein bound to this region in vivo ( Figure 3B ) and footprinted it in vitro ( Figure 3C and 3D ) . Second , transcription initiated at this promoter was phoP-dependent ( Figure 3E ) and extended into the phoP gene as determined by reverse transcription-PCR ( Figure 3F ) . Third , point mutations in the predicted PhoP box of the y1795 promoter abolished PhoP-dependent expression from this promoter ( Figure 3G ) . And fourth , Western blot analysis of cell extracts prepared from a Yersinia strain encoding a PhoP-HA protein from its normal chromosomal location and probed with anti-HA antibodies demonstrated higher levels of PhoP-HA protein following growth under inducing ( i . e . , low ) than in repressing ( i . e . , high ) Mg2+ concentrations ( Figure 3H ) . The incorporation of the y1795 gene into Yersinia appears to have occurred after the lineage that gave rise to this genus split from the one originating Serratia because y1795 orthologs have not been found outside Yersinia spp . ( see phylogeny in Figure 2 ) . The y1795 gene is predicted to encode a 207 amino acid outer membrane protein and/or lipoprotein without significant similarity to proteins of known function . That the y1795 gene is adjacent to the phoP gene and co-transcribed with phoP and phoQ suggests that it may affect the levels and/or activities of the Yersinia PhoP and PhoQ proteins . While this possibility cannot be presently ruled out , chromosomal replacement of the S . enterica phoP and phoQ genes with the Y . pestis orthologs ( without y1795 ) resulted in a strain that retained the normal regulation of the PhoP-activated mgtA gene [48] . Moreover , open reading frames harboring functions unrelated to those of the two-component system proteins sometimes precede two-component system genes . For instance , the phoPQ genes of Pseudomonas aeruginosa are preceded by and form an operon with oprH , which encodes an outer membrane protein [49] . Likewise , the genes for the PmrA/PmrB two-component system of S . enterica are part of a three-gene operon headed by the pmrC gene [50] , [51] , which encodes an inner membrane protein implicated in the modification of the lipopolysaccharide [52] . Mg2+ is the most abundant divalent cation in biological systems [53] . It is essential in the cytosol for ATP-mediated reactions and as a stabilizer of ribosomes [54] , and in membranes where it binds to negatively charged molecules , such as the phosphates in the lipopolysaccharide [55] . Consistent with low Mg2+ being a signal that activates the PhoP/PhoQ system [6] , several of the genes regulated by PhoP both in Yersinia and Salmonella and also present in the vast majority of the analyzed enteric species ( Figure 2 ) encode proteins mediating the adaptation to low Mg2+ . On the one hand , the PhoP protein directly promotes transcription of the Salmonella mgtA and mgtB genes [26] , encoding two of the three known Mg2+ transporters of S . enterica and exhibiting 50% identity to each other . PhoP directly regulates transcription of the mgtCB operon in Y . pestis ( Figure S2 ) , which lacks an mgtA gene . ( The mgtCB operon also encodes the inner membrane protein MgtC discussed above . ) Certain enteric species , such as E . coli and Citrobacter koseri , lack mgtCB but harbor mgtA ( Figure 2B ) ; other species such as Serratia proteamaculans and S . enterica harbor both mgtA and mgtCB whereas Sodalis glossinidius lacks homologs of all three genes . We identified putative PhoP boxes upstream of the coding regions of most of these Mg2+ transporter genes suggesting that the PhoP protein directly regulates their expression . On the other hand , the proteins encoded by the PhoP-activated pbgP operon and the ugd gene are responsible for covalently modifying the lipid A phosphates in the lipopolysaccharide with 4-amino-4-deoxy-L-arabinose at sites normally neutralized by Mg2+ , whereas the pagP gene product catalyzes the incorporation of a palmitate chain into lipid A [56] . These two modifications confer resistance to different antimicrobial peptides [57] . Except for Citrobacter koseri , all examined enteric species harbor pbgP and pagP homologs ( Figure 2 ) ; yet , as discussed above , homologs may not be functionally equivalent . Our analysis indicates that the PhoP regulon includes products that function in Mg2+ homeostasis , consistent with Mg2+ being the signal that regulates the PhoP/PhoQ system [6] . These products need not be highly conserved as long as they fulfill a required function . For example , E . coli and Yersinia harbor only one PhoP-activated Mg2+ transporter each , MgtA and MgtB , respectively , which are only 51% identical at the amino acid level , which is much lower than the 72% median identity that exists between E . coli and Yersinia proteins [19] . Only two loci are directly regulated by PhoP both in Salmonella and in Yersinia and present in all ten examined enteric species ( Figure 2 ) : phoP/phoQ , encoding the PhoP/PhoQ two-component regulatory system , and slyB ( Figure 4 ) , encoding an outer membrane lipoprotein with similarity to a lipoprotein implicated in membrane integrity in Burkholderia spp . [58] . Both phoPQ and slyB seem to be regulated by PhoP across the Enterobacteriaceae family because putative PhoP binding sites could be identified in their promoter regions in all analyzed enteric species ( Figures S3 and S4 ) . And in E . coli , direct PhoP regulation of phoPQ and slyB has been demonstrated experimentally [59] . Mutants lacking a functional phoP gene are defective for growth in low Mg2+ [6] , consistent with PhoP's role in governing the adaptation to low Mg2+ environments . We determined that a Salmonella mutant deleted for the PhoP box in the phoP promoter exhibited an identical phenotype ( Figure S5 ) . This demonstrates that positive autoregulation of the phoPQ operon is required to generate sufficient amounts of active PhoP protein to promote expression of the PhoP-regulated gene products mediating growth in low Mg2+ [6] , [26] . To explore the slyB function , we examined the behavior of strains lacking or overexpressing the slyB gene and experiencing low Mg2+ , which is a condition in which Salmonella requires a functional phoP gene [6] , [26] . Whereas a slyB-deleted strain grew like wild-type Salmonella , the strain overexpressing slyB was defective for growth , especially in low Mg2+ ( Figure 5A–B ) . To test whether the growth defect of the slyB-overexpressing strain was due to inhibition of PhoP protein activity , we examined transcription of four different PhoP-dependent promoters . We measured GFP expression levels in strains harboring transcriptional fusions to a promoterless gfp gene driven by either of four different PhoP-activated promoters in wild-type and slyB Salmonella strains grown under inducing conditions for the PhoP/PhoQ system . All promoters were expressed to higher levels in the slyB mutant compared to the isogenic wild-type strain ( Figure 5C ) . The slyB gene was responsible for this phenotype because the production of the SlyB protein from a plasmid restored the wild-type levels of expression of the PhoP-activated ugtL gene to a slyB mutant strain harboring a chromosomal lacZYA transcriptional fusion to ugtL ( Figure S6 ) . Together , these results indicate that slyB negatively regulates the activity of the PhoP protein . This effect appears to be specific to slyB because transcription of PhoP-dependent targets was not altered when eight other PhoP-activated genes were mutated ( data not shown ) . The ability of the PhoP/PhoQ system to modulate its own activity through positive and negative feedback loops has been conserved over hundreds of millions of years . The key role that these regulatory loops play in the proper functioning of the PhoP/PhoQ system is underscored by the fact that altering the activity of this system by either overexpressing the negative regulator slyB or abolishing phoPQ positive autoregulation render S . enterica unable to grow in low Mg2+ ( Figures 5A and S5 ) . The use of regulatory feedbacks to control the output of two-component systems is not exclusive to PhoP/PhoQ because other two-component regulatory systems positively regulate their own expression [60] and because the CseB/CseC two-component system of Streptomyces coelicolor controls the expression of a lipoprotein that negatively regulates CseB/CseC activity [61] . ( Note that even though the activity of the two-component systems CpxR/CpxA and RcsB/RcsC/RcsD is affected by the lipoproteins NlpE and RcsF , respectively , neither nlpE nor rcsF are controlled by the regulatory systems that they affect [62] , [63] . ) The growing number of available bacterial genome sequences has revealed the existence of large extents of genomic variability even among closely related bacterial species . Despite these differences , closely related organisms typically rely on orthologous regulatory proteins to respond to a given stimulus . Here , we have explored how the content of bacterial regulons is shaped in an environment of widespread gains and losses of genes .
Bacterial strains and plasmids used in this study are listed in Table S5 . Primers are listed in Table S6 . All S . enterica serovar Typhimurium strains were derived from wild-type strain 14028 s , and grown at 37°C in N-minimal medium [68] buffered in 50 mM Bis-Tris , pH 7 . 7 , supplemented with 0 . 1% casamino acids , 38 mM glycerol and 50 µM or 10 mM MgCl2 . Y . pestis strains were derived from wild-type strain KIM6 [69] , and grown at the optimal growth temperature of 28°C in defined medium [70] , pH 7 . 0 , supplemented with 0 . 1% casamino acids , 10 mM ( D ) -glucosamine , and 50 µM or 10 mM MgSO4 . E . coli strain DH5α was used as the host for the preparation of plasmid DNA . Ampicillin and kanamycin were used at 50 µg/ml and chloramphenicol at 20 µg/ml . After overnight culture in medium containing 10 mM MgSO4 , Y . pestis cells were washed with Mg2+-free medium and grown to A600 ∼0 . 3 in 16 ml of medium containing 50 µM MgSO4 with vigorous shaking . 14 ml of cell culture were collected , mixed with RNAprotect™ Bacteria Reagent ( Qiagen ) and used to prepare total RNA using RNeasy® Mini Kit ( Qiagen ) . RNA samples were treated with Turbo DNA-free DNase ( Ambion ) and re-purified with the RNeasy® Mini Kit . Y . pestis KIM tiling arrays were manufactured by NimbleGen Systems Inc ( Madison ) . The array features are illustrated on Figure S1 . RNA labeling , array hybridization and data extraction were carried out according to standard operating procedures by NimbleGen Systems Inc ( Madison ) . After overnight culture in medium containing 10 mM MgSO4 , Y . pestis cells were washed with Mg2+-free medium and grown to A600 ∼0 . 3 in 22 ml of medium containing 50 µM MgSO4 with vigorous shaking . ChIP assays were carried out essentially as described [71] . DNA labeling ( IP sample with Cy5; input DNA with Cy3 ) , array hybridization , data extraction and analysis were carried out according to standard operating procedures by NimbleGen Systems Inc ( Madison ) . Detailed information about these and other experimental protocols are provided in Text S1 . | Organisms often respond to environmental cues by modifying the patterns of expression of multiple genes . Related species typically rely on orthologous DNA-binding regulatory proteins to orchestrate the response to a given stimulus . However , it is unclear whether different organisms express similar or rather distinct groups of genes in response to the same signal . This is of special interest in bacteria because even closely related species , such as Escherichia coli and Salmonella enterica , have quite distinct lifestyles and display significant differences in gene content . Here , we have addressed this question by investigating the sets of genes that the Mg2+-responding PhoP protein controls in ten enteric species . We established that the majority of targets governed by the PhoP protein are not shared across species . This is due to regulation of species-specific genes and also to transcriptional rewiring of shared genes . The few PhoP-regulated targets retained across species encode the most conserved proteins , which determine the level of active PhoP protein and mediate Mg2+ homeostasis . These findings demonstrate that self-governing the activity of a regulatory protein is critical for its proper functioning , even though it may regulate largely different gene sets across species . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics"
] | 2009 | Evolution of a Bacterial Regulon Controlling Virulence and Mg2+ Homeostasis |
Mathematical modeling of behavioral sequences yields insight into the rules and mechanisms underlying sequence generation . Grooming in Drosophila melanogaster is characterized by repeated execution of distinct , stereotyped actions in variable order . Experiments demonstrate that , following stimulation by an irritant , grooming progresses gradually from an early phase dominated by anterior cleaning to a later phase with increased walking and posterior cleaning . We also observe that , at an intermediate temporal scale , there is a strong relationship between the amount of time spent performing body-directed grooming actions and leg-directed actions . We then develop a series of data-driven Markov models that isolate and identify the behavioral features governing transitions between individual grooming bouts . We identify action order as the primary driver of probabilistic , but non-random , syntax structure , as has previously been identified . Subsequent models incorporate grooming bout duration , which also contributes significantly to sequence structure . Our results show that , surprisingly , the syntactic rules underlying probabilistic grooming transitions possess action duration-dependent structure , suggesting that sensory input-independent mechanisms guide grooming behavior at short time scales . Finally , the inclusion of a simple rule that modifies grooming transition probabilities over time yields a generative model that recapitulates the key features of observed grooming sequences at several time scales . These discoveries suggest that sensory input guides action selection by modulating internally generated dynamics . Additionally , the discovery of these principles governing grooming in D . melanogaster demonstrates the utility of incorporating temporal information when characterizing the syntax of behavioral sequences .
Sequential animal behaviors are often composed of repetitions of simple subroutines . In driving motor action execution , nervous systems must integrate sensory information with internal priorities and dynamics . Both external and internal conditions contribute to sequence organization , but their respective weights are unknown . At one extreme , purely sensory-driven , reflexive processes produce actions , such as larval escape sequences , solely based on recent sensory input [1] . At the other , internally generated and maintained nervous system dynamics , such as those found in the crustacean stomatogastric ganglion , may produce behaviors which proceed irrespective of changing external conditions [2 , 3] . Grooming , a common behavior across species , confers social and survival benefits [4 , 5] and provides a rich source of data for discovering rules that organisms use to produce behavioral sequences . The vocabulary of grooming consists of the possible actions that may be executed and has been cataloged in mice and several species of flies [6–8] . However , the rules for grooming action sequence organization , or syntax , remain poorly understood . In Drosophila melanogaster , or fruit flies , these sequences are variable between individuals and within individuals across grooming sessions , suggesting that flies use non-deterministic rules to make grooming decisions . Because of the complex structure of these sequences , analysis of D . melanogaster grooming can reveal distinct rules of sequence generation at different temporal scales: the long time scale of bulk behavioral progression , the intermediate scale of grooming motifs , and the short scale of individual grooming bouts . On long time scales , D . melanogaster grooming in response to exposure to an irritant is well-described as a process that typically progresses from an early phase characterized by preferential grooming of anterior body parts , such as eyes , to a later phase which exhibits heightened proportions of walking , abdomen cleaning , and wing cleaning . This progression occurs gradually over the course of many minutes . Previously published computational models which utilize either hierarchically structured suppression or graded sensory gain exhibit gross features of this progression , namely a gradual transition from early anterior-heavy grooming to a later quasi-steady state featuring increased posterior grooming and walking levels [9] . At an intermediate temporal scale , grooming is organized in units that we refer to as motifs . We identify two classes of motifs ( anterior and posterior ) , which are named for the set of legs used to execute them ( front and back , respectively ) . Motifs consist of consecutive alternations between body-directed grooming bouts and leg rubbing bouts . Bouts are defined as sustained periods of a single grooming action ( e . g . head cleaning or wing cleaning ) . Grooming bouts occur at the shortest time scale we consider here , as individual bouts typically last somewhere between 150 ms and 2 s . Flies use the same pair of legs to execute bouts within a motif , allowing them to transition between within-motif actions easily . Specifically , an anterior grooming motif consists of consecutive alternations between bouts of head cleaning and front leg rubbing . A posterior motif consists of alternating bouts of abdomen cleaning , wing cleaning , and back leg rubbing . Analysis of freely-behaving flies reveals behavioral structure at multiple time scales , with coarse-grained descriptions being sufficient to describe long-term trends and higher resolution descriptions providing increased predictive power at shorter time scales [10] . This finding suggests that there is value in examining sequential data at several levels of temporal resolution . Additionally , it suggests that treating behaviors as both continuously varying and discretely separated in the same analysis can yield more insight than considering either purely continuous or discrete models . Currently , we do not know how flies integrate sensory information with internal states in order to produce multi-scale grooming behavior . It is possible that long time scale grooming trends are governed principally by changing sensory conditions , as grooming results in the removal of irritant over time . Several studies also provide evidence that sensory input is sufficient to drive grooming behavior in a reflex-like fashion at short time scales [11 , 12] . Here , we use statistical models to characterize grooming syntax at each of the temporal scales mentioned above . Several classes of Markov models , in which the probability of an event occurring is contingent upon previous events , have been used to describe factors involved in non-deterministic decision-making in various animals . Different Markov models vary in their structure and number of parameters but they each require a well-defined state space . Navigation and foraging have been described using Markov models , as these sequential behaviors can be decomposed into subroutines which can be categorized using easily observable dimensions such as direction and velocity [13–15] . These models are applicable to common behaviors exhibited by D . melanogaster as well , such as courtship and locomotion [16–18] . Recent work from Tao et al . uses a Hierarchical Hidden Markov Model ( HHMM ) to analyze fruit fly locomotor behavior in the presence of odors [19] . They find that , although locomotor behavior is non-stereotyped , it can be decomposed into stereotyped units , making it suitable for analysis within a Markovian framework . Berman et al . [20] use high-order Markov models to describe the behavior of freely-roaming D . melanogaster , suggesting the applicability of such models to behavioral subsets identified in their analysis , including grooming behavior . It should be noted that non-Markovian dynamics have also been identified in animal vocal sequence production , suggesting that other classes of models may be useful for describing behavioral generation [21–23] . In fact , Berman et al . [20] report that , although a Markovian framework is useful for illustrating some features of behavioral transitions , they observe long time dependencies in their data that cannot be captured by their Markov models . Here , we use our in-house Automatic Behavior Recognition System ( ABRS , https://github . com/AutomaticBehaviorRecognitionSystem/ABRS/ ) , which can classify different grooming movements from videos of flies covered in dust , to generate a large data set of ethograms—records of cleaning actions over time—from wild-type flies removing dust . We analyze more than 40 total hours of video from over 90 flies ( Fig 1 ) . First , this data set allows us to quantify behavioral trends at several temporal scales on a larger data set than has been previously described . Fig 2 provides a schematic overview of the temporal scales we analyze here . Notably , we observe the progression from anterior to posterior grooming that has been identified by Seeds et al . [9] . We also observe a strong correlation between body-directed and leg-directed cleaning actions within motifs . That is , the amount of head cleaning and front leg rubbing are extremely strongly correlated over the entire course of grooming , as are the amount of back leg rubbing and the sum of abdomen and wing cleaning . Finally , we observe a relaxation into steady state-like behavior after approximately 13 minutes of grooming . Next , we use a set of Markov models to characterize probabilistic rules governing grooming action transitions . We use a Markovian framework as a first approximation in order to discover relevant features of grooming , since Markov models are simple and carry few assumptions . As such , these models serve as a tool for grooming sequence feature identification rather than as a full explication of grooming decision-making factors . Previous work has indicated that such analysis can reveal temporal relationships in sequential behavioral data , indicating that , even if the behavior being analyzed is not fully Markovian in reality , Markov models can still be useful as a descriptive exploratory tool [16–20] . The simplest of our models , a first order discrete time Markov chain , highlights the presence of grooming motif structure . Subsequent models incorporate temporal information by considering grooming bout durations . By comparing these models to statistical null model hypotheses , which shuffle bout order or bout duration , we discover the contribution of bout duration to grooming sequence structure at short time scales . Though we rely on a Markovian framework to identify features of grooming syntax , our analysis does not preclude the possibility of non-Markovian dynamics in fruit fly grooming . Indeed , our data exhibits two notable nonstationarities . First , we observe an overall trend in grooming proportions , as has been reported previously [9] , wherein flies favor anterior grooming immediately after being dusted but favor posterior grooming and walking after approximately 10-15 minutes . Second , grooming transition probabilities change slightly over the course of grooming ( though they are unexpectedly stationary over time ) , indicating that a purely stationary model is an oversimplification . To account for this , we extend our analysis to include a simple nonstationarity which captures sequence progression more closely than a first order Markov chain . We find that a time-varying Markov renewal process ( MRP ) which incorporates bout duration dependence recapitulates the observed grooming structure at long , intermediate , and short time scales . This model is partially Markovian , as transition probability matrices dictate state transition dynamics . However , the model also includes renewal process dynamics , in which the duration of the ensuing bout is determined using a random draw from the empirically observed bout duration distribution . This model is nonstationary due to the fact that Markov transition probabilities change over time . This model recapitulates both the observed bulk grooming progression statistics and the fine-grain temporal structure of grooming motifs , suggesting that sensory drive can modulate internal programs in a simple manner to prioritize different grooming strategies in different contexts . Our models clarify the syntax of grooming , demonstrating that , at the scale of individual bouts , transitions are dictated by both the identity and duration of the preceding bout . While previous studies have established the non-random nature of fruit fly grooming [6 , 9 , 16–18 , 20] , this work reveals the previously unknown contribution of grooming bout duration to sequence structure , suggesting a role for sensory input-independent decision-making . Finally , a model that includes duration dependence at the scale of individual grooming decisions and a linear progression to mimic dust removal is able to generate synthetic ethograms that closely resemble observed data at the intermediate scale of motifs and the long scale of the grooming progression . The discovery of duration dependence in grooming sequence structure suggests the existence of neural circuits in D . melanogaster that can generate patterned behavior irrespective of sensory drive , as transition rules and bout durations remain nearly constant even as the dust distribution varies over the course of grooming . Additionally , the tight correlation between body and leg-directed actions within motifs demonstrates for the first time the existence of duration dependence at an intermediate temporal scale . Finally , the model improvements provided by the inclusion of a simple nonstationarity suggest that sensory drive plays a role in guiding the implementation of internal programs over long time scales rather than serving a purely reflexive action selection function at short time scales . These discoveries provide the groundwork for future experiments and analysis of the neural underpinnings of duration-dependent action selection and sensory-driven modulation of decision-making in grooming and related behaviors .
Drosophila melanogaster ( N = 92 ) were recorded at 30 Hz for 27 . 8 minutes after being uniformly covered in Reactive Yellow 86 dust ( Organic Dyestuffs Corporation , Concord , North Carolina ) . To generate labeled ethograms , we apply the ABRS classifier to video ( https://github . com/AutomaticBehaviorRecognitionSystem/ABRS ) . Following this , real time ethograms are converted to discrete bout ethograms and de-noised ( Fig 1 ) . Briefly , the classifier is trained on grooming data using a combination of supervised and unsupervised protocols in several steps . First , a Fourier transform is performed on pixel intensity data across 17-frame sliding windows ( ∼0 . 5 s ) , resulting in two-dimensional spectra referred to as space-time images ( ST-images ) . ST-images are subsequently modified by Radon transformation , which can be used to represent 2D shapes in a rotation and translation-invariant manner . Fast Fourier transformation is then applied to the Radon transformed ST-images , resulting in position and orientation-invariant ST-images [24] . ST-images are then decomposed using singular value decomposition ( SVD ) and clustered using linear discrimination analysis ( LDA ) with human labeled behavioral categories . Clusters identified using LDA correspond to five predefined grooming behaviors ( f = front leg rubbing , h = head cleaning , a = abdomen cleaning , b = back leg rubbing , w = wing cleaning ) and two non-grooming behaviors ( wk = walking , s = standing ) . After classification , each video is represented by a 50 , 000-element vector , with each element containing a label denoting the behavior identified in the corresponding frame . Classifier performance and validation are shown in S1 Fig . To generate discrete bout ethograms , each 50 , 000-element vector is transformed into a two-dimensional vector in a series of steps . The first row of this discrete bout ethogram contains the bout identity , or the label corresponding to one of the seven behaviors described above . The second row contains the bout durations , or the amount of time spent in the corresponding behavior before transitioning to another behavior . Consecutive frames containing identical actions are consolidated into one bout identity entry . Then , the corresponding bout duration entry is calculated by counting the number of consecutive identical frames and converted into seconds by dividing by the video frame rate ( 30 Hz ) . As a result , the number of entries in discrete bout ethograms corresponds to the number of observed bouts , which varies between flies . However , the sum of bout durations for each ethogram is identical ( 27 . 8 minutes ) . After ethogram generation , behavioral bouts persisting for less than 167 ms are deemed artifacts and deleted . This threshold is chosen because manual inspection of video data suggests that individual leg sweeps occur at approximately this time scale . This is also the approximate average transition time between behavioral states described by Berman et al . [10] . Deleted bouts account for 1 . 03% of total recording time . In subsequent analysis , we also neglect standing bouts , as they account for < 1% of total recording time . In order to include bout duration as a data dimension , we introduce a binning scheme . To implement this change , the continuous-valued entries in the second row of discrete bout ethograms are collapsed into duration category labels . Grooming bouts are classified into three categories ( short , medium , and long ) based on the duration of the associated bout . Although our results are consistent even when using more than three duration categories ( S2 Fig ) , we limit analysis to three categories for the sake of simplicity . Bin edges for these categories are determined independently for each grooming action ( f , h , a , b , w ) and walking ( wk ) , as each action possesses a unique duration distribution . Bin edges are chosen such that each bin is populated by an equal number of samples ( i . e . the total number of actions labeled “short” is equal to the number of “long” actions ) and are depicted in Fig 3 . Markov processes are a class of probabilistic models that describe transitions between states of a system in terms of previous states . The simplest version obeys the Markov property , which declares that the next transition between states depends only on the current state . As such , long-term history dependence can be neglected when considering Markov processes , greatly simplifying the models . In discrete time , the dynamics of a first order Markov chain are described by the equation x t + 1 = x t M , ( 1 ) where xt denotes the state probability vector at time t . The entries of xt represent the probability of the system being in the state corresponding to that entry at time t . M is the matrix of state transition probabilities such that entry Mij is the probability of transitioning from state i to state j . The vector xt will have dimensionality corresponding to the number of possible states in the system . In this case , there are either six , twelve , or eighteen grooming states , depending on whether grooming bout duration is incorporated into the model . Given sufficient data about state transitions in a real system , it is possible to generate a maximum likelihood estimate ( MLE ) transition probability matrix , denoted by M ^ , which provides the best estimate of the transition probabilities governing the observed transitions . The entries of M ^ are given by M ^ ij = n ij ∑ u = 1 v n iu , ( 2 ) where nij is the total number of observed transitions from state i to state j . The summation over niu is a normalization factor which counts the total number of transitions from state i to any other state , indexed by u , and v denotes the total number of possible states . First order Markov chain maximum likelihood transition probabilities are determined for each ethogram as described in Eq ( 2 ) . Probabilities are determined for ethograms with either one , two , or three bout duration bins . The resulting matrices represent population-wide average transition probabilities . We also calculated transition matrices for individual flies , examples of which are shown in S3 Fig . Statistical null models , which randomize isolated features of data while maintaining overall statistical features , provide a basis for hypothesis testing to determine the significance of results obtained from experimental data . To assess the contributions of bout identity and bout duration to sequence structure , we generate null hypotheses which independently randomize these features of the ethogram data while preserving bout duration frequency distributions within grooming action types . Null hypothesis transition probabilities are analytically tractable in the limit of infinite permutations , so we use exact formulas to calculate matrix entries ( S1 Table ) . A representative example of each permutation on grooming bout order and duration is shown schematically in Fig 4 . Our null hypotheses are as follows: The resulting model fit is determined using the Bayesian Information Criterion ( BIC ) , defined as BIC = log ( n ) k - 2 log ( L ) , ( 3 ) where n is the number of data points observed , k is the number of free model parameters , and log ( L ) is the maximized log-likelihood value . For first order Markov chains , log ( L ) = ∑ t = 1 T - 1 P ( x t + 1 | x t ) , ( 4 ) and P ( x t + 1 | x t ) = M ^ ij , ( 5 ) where i is the row index for the behavior observed at time t and j is the column index for the behavior observed at time t+1 . The BIC provides a metric for evaluating model goodness-of-fit by penalizing models which use a large number of free parameters ( the first term in Eq 3 ) and rewarding models which accurately predict unobserved data ( the second term in Eq 3 ) . Thus , models with low BIC values possess more explanatory power than models with higher BIC values but still avoid overfitting . BIC is similar to a related metric , Akaike’s Information Criterion ( AIC ) , but penalizes models with more free parameters more severely than AIC in cases where the number of data points is much larger than the number of free parameters . To evaluate the ability of our models to generate sequences that are similar to our observed data , we use them to create synthetic ethograms . Synthetic sequences are generated using Monte Carlo methods within a Markov renewal process ( MRP ) framework . In an MRP , state transitions are governed in a manner identical to Markov chains ( i . e . using a transition probability matrix ) . However , the amount of time spent in each state is determined in parallel by a renewal process , in which action durations are drawn from a known distribution . Here , we use the empirically observed duration distributions shown in Fig 3 .
After exposure to an irritant , D . melanogaster grooming begins with an initial phase of mostly front leg rubbing and head cleaning ( f , h ) . We refer to these behaviors as anterior motif grooming actions . After roughly 13 minutes , flies reach an approximate steady state consisting of heightened amounts of abdomen cleaning , back leg rubbing , and wing cleaning ( a , b , w ) . Together , these actions constitute the posterior motif . Flies also exhibit an increased proportion of walking ( wk ) during the late phase of grooming . Our result , visualized in Fig 5 corroborates a similar finding reported by Seeds et al . [9] but utilizes a much larger , automatically annotated data set . We find that grooming bouts possess action-specific duration distributions . Fig 3 depicts this result . We find that anterior motif grooming actions possess remarkably similarly shaped action duration profiles , with peaks in their probability functions near 1 s . In contrast , posterior motif actions tend to be shorter . Most notably , abdomen and wing cleaning exhibit sharp distribution with peaks near 250 ms , whereas back leg rubbing exhibits a smoother distribution which more closely resembles those of anterior motif actions . Interestingly , bouts durations are not normally distributed , suggesting a mechanism other than random generation of durations around a mean value . Instead , anterior motif duration distributions possess “shoulders” on each side of the central peak , suggesting that dividing actions into bout duration categories is not an unnatural distinction . Moreover , these “shoulders” are relatively broad due to the fact that grooming bouts consist of many quanta of individual leg sweeps ( in the case of body-directed actions ) or rubs ( in the case of leg-directed actions ) , which are cyclical in nature and last approximately 150 ms per cycle . Due to video frame rate resolution limitations , we do not consider this extremely fast behavioral scale here . In this analysis , we highlight results which divide grooming actions into three duration categories in order to provide an intuitive demarcation between “short” , “medium” , and “long” actions . However , it is important to note that our conclusions are not strictly dependent on this choice of categorization scheme , as our results hold for higher numbers of duration categories as well S2 Fig . We find that the vast majority of transitions occur within grooming motifs . Approximately 88% of transitions from anterior motif grooming bouts are to another anterior motif behavior , while about 70% of transitions from posterior motif bouts are to another posterior motif behavior . This tight intra-motif coupling is illustrated in Fig 5 , as the amount of front leg rubbing tracks the amount of head cleaning extremely closely . The amount of back leg rubbing also tracks the sum of abdomen and wing cleaning very closely . This result is not affected by the choice of sliding window , as we observe a similarly strong correlation even when using a 1 . 67 s ( 50 frame ) window instead of the 16 . 7 s ( 500 frame ) window used to illustrate the progression shown in Fig 5 . We find that this strong coupling emerges from the structure of individual grooming motifs . Fig 6 provides an illustrated example of the structure of an anterior motif . We find that , across flies and independent of the time after grooming , anterior motifs contain nearly perfectly correlated amounts of body and leg-directed grooming . This strong correlation is somewhat surprising given that consecutive anterior grooming bouts exhibit a non-linear relationship ( S4 Fig ) . This tight intra-motif coupling is also reflected in the structure of Markov transition matrices fit to our behavioral data ( ethograms ) . We fit a first order discrete time Markov chain to our data and the resulting matrix , denoted by M ^ ( Eq 2 ) , contains the maximum likelihood estimates of the probabilities of transitioning from one grooming action to another . Fig 7 depicts the structure of these transitions in both network and matrix form , illustrating the transition probabilities used to define the Markov chain in Eq ( 1 ) . In Fig 7 , anterior and posterior motif grooming actions are indicated by light red and blue outlines , respectively . Note that self-transitions are explicitly prohibited as a result of the ethogram discretization method . Consequently , the network diagram lacks self-loops and all diagonal elements in transition matrices are blacked out , signifying zero value entries . In Fig 7 , we illustrate the fine-grain temporal structure present within the identified motifs . Transition matrices with either two ( left ) or three ( right ) duration bins reveal the presence of certain high-probability transitions . Most notably , we observe that transitions between long anterior grooming actions dominate to an extent not found in the null hypotheses we use for comparison . For example , when using three duration categories , the probability of transitioning from a long front leg rub to a long head cleaning bout is ∼39% . This is significantly higher than the corresponding transition probabilities in the duration permuted and order permuted null hypotheses , which are ∼28% and ∼13% respectively . These differences are depicted in greater detail in Fig 8 . Additionally , we observe that , regardless of the number of duration bins , the anterior motif transition probability structure is nearly mirror symmetrical along the diagonal ( i . e . transitions between front leg rubbing and head cleaning bouts have the same probabilities ) . In contrast , posterior motif transitions are less strongly symmetrical , with transitions between abdomen cleaning and back leg rubbing exhibiting the highest probabilities ( Fig 7 ) . We find that the probabilistic rules describing transitions between consecutive bouts depend on two features of immediate behavioral history: previous bout identity and previous bout duration . To validate this observation , we evaluate the quality of the maximum likelihood model using the Bayesian Information Criterion ( BIC ) . This metric , which is described mathematically by Eq 3 , rewards models that produce accurate predictions and penalizes those that contain more parameters . The model with the lowest BIC score is considered the best fit to the observed data . Using BIC , we show that , compared to statistical null model hypotheses , the maximum likelihood transition probability best describes the data without overfitting . BIC values and null hypothesis matrices used for comparison are shown in Fig 8 . Since each null hypothesis matrix contains fewer free parameters than the maximum likelihood model , the advantage of the maximum likelihood matrix in BIC is due entirely to its superior explanatory power . Inspection of BIC values indicates that bout order is the strongest individual determinant of grooming syntax and that bout duration provides an additional contribution . Fig 8 shows the BIC values for each matrix as determined using data binned into three duration categories . The order permuted null hypothesis possesses a significantly larger BIC value than the duration permuted null hypothesis , indicating that disrupting bout order degrades predictive power most significantly . By construction , each null hypothesis transition matrix contains identical row structure within sets of rows corresponding to the same grooming action . However , only the duration permuted transition matrix possesses block structure similar to that found in the maximum likelihood matrix , as seen in Fig 8 . This feature also reinforces the claim that action order provides a relatively larger contribution to syntax , since preserving action order results in the preservation of transition matrix block structure . Fig 9 shows the average transition probability matrices with three duration categories for the early and late phases of grooming , as separated by the dashed line in Fig 5 . Data were separated into early and late phase data for each fly and maximum likelihood transition probabilities were determined as described in Methods and Models . Here , we determine the boundary between phases to be the average time elapsed after flies have performed half of their discrete grooming actions . We also conducted a similar analysis in which we fit transition matrices only to the first and last third of data so as to avoid using the middle portion of the progression . This did not alter our results ( S5 Fig ) . Qualitatively , these matrices exhibit strong similarities , despite the differences in grooming proportions , as shown in Fig 5 . In particular , transitions between anterior motif grooming movements are similar across phases of the recording . Additionally , we find that 253 of the 266 ( 95 . 1% ) non-zero entries change by less than 5% and 168 ( 63 . 2% ) entries change by less than 1% ( S6 Fig ) . This suggests that the rules for sequence generation are nearly fixed , in spite of changing sensory input . Most notably , transitions between long abdomen grooming bouts and long back leg rubbing bouts are enriched in the late phase of grooming , illustrating that grooming rules , while similar , are not completely stationary over time . Markov models are steady state models and , as such , only exhibit equilibrium dynamics when used in a generative manner to create synthetic data . The first order maximum likelihood Markov model presented above is best suited for describing the late stages of grooming , which resembles steady state behavior . This is an obvious shortcoming of using a Markovian framework for analyzing the ethograms presented here , as flies display a dynamic progression ( Fig 5 ) , likely due to changing amounts of irritant over the course of grooming . In order to more explicitly model these changing sensory dynamics , we utilize a nonstationary , or time-varying , Markov renewal process ( MRP ) . Here , we introduce a nonstationarity in the form of a time-varying transition probability matrix . In order to maintain as parsimonious a model as possible , we define the transition probability matrix as a time-dependent convex combination of an early phase matrix and a late phase matrix ( Eq 6 ) . The early phase matrix is determined using maximum likelihood fitting on only the first 200 actions of each fly . Likewise , the late matrix uses each fly’s final 200 actions . This ensures that early and late phases are well-separated , as each ethogram averages approximately 1000 bouts . The time-varying transition probability matrix M ( t ) used in the MRP is defined as M ( t ) = { 13 - t 13 M e a r l y + t 13 M l a t e , t < = 13 , M l a t e , t > 13 ( 6 ) where t is in minutes . After 13 minutes , the late phase transition matrix is used due to the observation that flies exhibit steady state behavior after approximately that amount of time . Bout duration distributions used in the MRP do not change over time , as we did not find any significant difference between early and late bout durations ( S7 Fig ) . Surprisingly , using this simple linear interpolation as a proxy for changing sensory conditions yields a model which produces ethograms that closely resemble our observed data ( Fig 10 ) . This interpolation serves as a first-order approximation to changing sensory drive , as flies constantly remove dust at an unknown rate over the course of grooming . Synthetic flies prioritize anterior grooming early on and gradually transition into a late phase which prioritizes posterior grooming and walking . This synthetic progression closely matches the actual progression shown in Fig 5 . The synthetic ethograms also exhibit a tight coupling between the amount of anterior grooming actions over time . Moreover , synthetic motifs possess similar structure to observed motifs , as shown in panel C of Fig 10 . Finally , the use of empirical transition matrices and bout duration distributions in our MRP guarantees that synthetic ethograms possess the same duration dependence at short time scales as actual data .
After exposure to an irritant , D . melanogaster engage in a series of grooming actions in order to clear the irritant from their body . These movements are highly non-random , as indicated by their structure at three separate temporal scales . On the longest time scale , grooming progresses over several minutes from an early anterior-heavy phase to a later phase which contains elevated levels of posterior-directed actions and walking as illustrated in Fig 5 . At an intermediate time scale , anterior and posterior grooming motifs exhibit strong correlations between the amount of body and leg-directed movements present in each motif , as shown in Fig 6 . Finally , grooming syntax exhibits non-random structure at the short time scale of individual grooming bout decisions , as illustrated by the non-random structure of transition matrices shown in Fig 7 . Together , these findings suggest that there are distinct syntactic rules that govern grooming sequence structure at different temporal scales . Here , we isolate short time scale grooming syntax by using a binning scheme to represent grooming bouts as semi-discrete actions ( e . g . short , medium , or long bouts ) . Across different binning schemes , within-motif transitions ( i . e . anterior-to-anterior or posterior-to-posterior ) dominate syntax at the scale of consecutive bouts , as indicated by the block-like structure of transition matrices in Fig 7 . The likely functional explanation for this tight correlation is that D . melanogaster engage in leg-centric grooming as a way to clear irritant that has collected on their limbs after body-centric grooming . Past work on blowflies also suggests that this coupling may be due to postural considerations , since fewer motor actions are required to transition between movements which use the same set of legs [6] . Moreover , the amount of front leg rubbing over the course of the grooming progression corresponds nearly exactly to the amount of head cleaning performed by the flies . The amount of back leg rubbing is also directly proportional to the amount of abdomen and wing cleaning ( Fig 5 ) . Though the total amount of time spent performing actions within motifs is tightly coupled , consecutive bout durations do not exhibit a simple linear relationship ( S4 Fig ) . Instead , transitions between body-centric and leg-centric actions of certain durations dominate . In particular , anterior motif movements exhibit nearly symmetric dynamics , with a preference for long-to-long transitions . In contrast , posterior motif transitions are asymmetric—this can be seen in Fig 7 , as transitions from body-directed actions to leg-directed actions have different values than transitions in the opposite direction . This suggests a mechanism whereby internally generated dynamics that are specific to each motif guide intra-motif transitions . Anterior motif transitions display symmetric transition rules , as shown in Fig 7 . The transition from head cleaning to leg rubbing may exhibit a strong correlation in duration for the following reason—the longer the head cleaning bout , the more particulate the legs accumulate , which subsequently requires more time to clear . However , it is more surprising that after performing a bout of leg rubbing the animal should display a similar duration preference upon transitioning back to head cleaning; once the front legs are sufficiently clean , we would not expect the animal to exhibit any duration preference on the subsequent bout . Additionally , the bout duration distributions for grooming actions remain strikingly similar across the entire recording , suggesting that sensory considerations do not account for the duration dependence we observe ( S7 Fig ) . These observations support the hypothesis that alternations between anterior grooming bouts are regulated by a common source or common dynamics that dictate intra-motif transition frequency . If this is the case , future investigation of neural circuitry could focus on identifying neural activity that oscillates in phase with grooming bout alternations . Under this model , flies would then need to combine discrete decisions about which motif to perform with continuous decisions about how long to maintain a motif . This provides other targets for future experiments , as we could look for neural activity corresponding to transitions between motifs , explore manipulations that induce transitions between motifs , and search for factors that affect the duration of motifs . Although sensory input is sufficient to initiate individual grooming bouts , it has been difficult to assess the role of sensory input in grooming on longer time scales , since direct quantification of irritant relies on invasive protocols [25] . To date , the contribution of internal dynamics to grooming sequence generation also remains unknown . The discovery of duration dependence in grooming suggests that flies do not rely exclusively on sensory information to make grooming decisions , leaving open the possibility that sensory input modifies pre-existing autonomous grooming programs . Here , we abstract the contribution of sensory input , incorporating it as a parameter in a time-varying Markov renewal process . Specifically , we approximate changing sensory conditions as a linear phenomenon which dictates the degree to which flies favor early versus late grooming rules ( Fig 10 ) . This choice of nonstationarity preserves the parsimony of our MRP , as it precludes the necessity of re-calculating transition probabilities over time . Additionally , the similarity between early and late intra-motif transition probabilities ( Fig 9 ) provides some leeway for the choice of nonstationarity , bolstering our confidence that a simple linear rule is sufficient to approximate true sensory conditions . Since our MRP reliably reproduces true ethogram statistics without explicitly modeling sensory input , we propose that , rather than acting in a purely reflexive manner , D . melanogaster instead use sensory information to modulate internal grooming programs . This type of slowly varying process has been observed in other organisms , suggesting that it is a useful framework for describing sequential behaviors [26] . Additionally , this hypothesis can be tested in future experiments which manipulate sensory conditions using irritants with different properties or optogenetic stimulation of sensory receptors . From a computational perspective , sensing is costly when it requires high-frequency updates and can provide readouts at a high resolution , as specialized neural circuits must be formed , maintained , and integrated in order to provide accurate , real-time readouts . This can lead to increased metabolic costs on several fronts , as increased neural activity and behaviors intended to protect and maintain sensory machinery can also require energetic resources [27] . In contrast , autonomous internal dynamics require only sparse sensory updates and may not require the same level of neural activity as is needed for sensory processing . If sensory drive alone , triggered by the presence of an irritant , dictates sequence generation in a purely reflexive manner , competing priorities may lead to rapid alternations between actions without any distinctive duration structure , consistent with neural circuity implementing competitive inhibition with a transition cost [9] . During the execution of a grooming motif , frequent , high-resolution sensory updates may not be necessary . Instead , the moment-to-moment decisions about execution of grooming bouts can be automated by utilizing patterned internal dynamics . Several models of behavioral generation in which internal triggers drive sequence generation have been proposed [28 , 29] . Since we show that intra-motif transitions exhibit bout duration-dependent structure , we suspect that internal triggers may provide a basis for transitions on short time scales after a discrete grooming motif decision is made . We consider this a strong possibility due to the fact that anterior motif transition probabilities are nearly stationary across grooming even though the dust distribution on the body is constantly changing . The near stationarity of these rules in spite of a changing stimulus suggests to us that behavior possesses an internally generated , fixed component . The discovery that grooming progresses gradually from anterior to posterior movements suggests that , on long time scales , D . melanogaster utilize sensory information to dictate decisions about which grooming motif to perform , since the dust distribution changes over the period of cleaning . In discussing temporal correlations between neural activity across spatial scales , Berman et al . [10] note that “ ( a ) lthough no such correlation has been specifically found in Drosophila , our results suggest that such neuronal patterns may exist: perhaps by combining descending commands from the brain with local circuitry within and emerging from the ventral nerve cord . ” This observation is consistent with our interpretation of our results , as local circuitry in the ventral nerve cord could plausibly generate internal dynamics which introduce duration dependence to action selection . By combining sensory input and internal dynamics , Drosophila melanogaster nervous systems may utilize multi-level control algorithms which make discrete , “ballistic” decisions about the onset and type of continuous behaviors , allowing them to update infrequently at low sensory resolution . Once the decision is made , the execution of the behavior can be guided primarily by autonomous internal dynamics . In flies , these internal dynamics may not be unique to grooming , as freely-behaving D . melanogaster exhibit hierarchical Markov-like behavior in the absence of external sensory stimuli [10 , 18] . Additionally , the identification of stereotyped subroutines in D . melanogaster locomotion further supports the idea that specialized neural circuits can automate and reproduce portions of behavior , reducing the need for constant calibration via sensory feedback [19] . We can further explore this idea by studying whether spontaneously grooming flies or those stimulated using optogenetic methods exhibit similar syntactic rules . In larval and mature D . melanogaster , specialized , isolated neural circuits known as central pattern generators ( CPGs ) , use rhythmic activity to guide execution of locomotion and courtship song sequences [30 , 31] . In humans , CPGs are believed to facilitate chewing and breathing subroutines , such as individual jaw or pharyngeal movements [32 , 33] , indicating that such neural circuit elements are common and flexible enough to carry out a wide variety of functions . Since CPGs can modulate motor programs independently of sensory input , they are a natural candidate to execute grooming behavior on the scale of individual leg sweeps and rubs within a grooming bout . We do not explicitly model behavior at this temporal scale , but the ABRS classifier applied to video at the frame rate used here identifies actions nearly at the temporal resolution of individual leg movements . We plan to extend analysis to this time scale in future work . Additionally , robotic systems that utilize multi-layer CPG architectures can carry out locomotor tasks , indicating that hierarchical CPG structures can generate sequential behaviors from smaller elements [34] . Recent work indicates that circadian rhythms and genetic factors contribute to D . melanogaster grooming behavior as well , leaving open a role for internal regulation of sequence generation at longer time scales [35] . When describing complex behavioral phenomena , formal mathematical models can bridge the gap between phenomenology and biological mechanisms by providing parameters that may correspond to underlying neuronal activity . In many cases , statistical model parameters do not directly describe to underlying biological structures , but data analysis can suggest hypotheses for future exploration . Here , we analyze large-scale automatically annotated data sets to reveal the syntax of D . melanogaster grooming . Using Markov models to detect features of grooming sequences , we find that grooming actions exhibit duration dependence at the scale of individual bouts and grooming motifs . Then , we produce realistic synthetic ethograms by introducing a slow modulation of grooming rules meant to abstract the contribution of sensory input . Together , these findings suggest that internal programs dictate grooming decision on fast time scales but are modulated by sensory input over longer time scales . As the tools for computational analysis of behavioral data continue to develop , interdisciplinary approaches that use mathematical tools to illuminate biological mechanisms will yield further insight into the generation of sequential behavior . Future experiments involving the manipulation of sensory experiences using optogenetic stimulation or other methods will also allow researchers to test hypotheses with unprecedented precision and scope . Together , these advances promise to improve our understanding of sequence generation by connecting mathematical descriptions of behavior with the underlying neural circuitry , as we are now motivated to search for circuits controlling long , intermediate , and short time scale grooming rules and determine how they interact with sensory circuits . Finally , our work suggests that temporal dynamics can and should be included when using statistical models to assess complex behaviors . | Analysis of temporally rich behavioral sequences provides a quantitative description of the rules underlying their generation . Drosophila melanogaster grooming behavior consists of many complex sequences involving repetitions of well-characterized actions . In this paper , we leverage advances in machine vision to automatically annotate over 40 hours of video data of flies covered in dust and develop mathematical models that reveal the existence of syntax in D . melanogaster grooming . We find that sequence organization depends on grooming action identity , as has been well-established , and , more surprisingly , grooming action duration . The discovery of duration-dependent action selection leads us to conclude that , although sensory input informs grooming decisions on long time scales , internal dynamics also guide individual transitions between grooming actions . Therefore , incorporating action duration into our models allows us to uncover multi-scale temporal dynamics that suggest the existence of neural circuits dedicated to partially sensory-independent decision-making . Our approach highlights the importance of incorporating temporal information into sequential models , as doing so reveals the relative contributions of sensory input and internal dynamics to behavioral sequence generation . | [
"Abstract",
"Introduction",
"Methods",
"and",
"models",
"Results",
"Discussion"
] | [
"invertebrates",
"linguistics",
"medicine",
"and",
"health",
"sciences",
"abdomen",
"markov",
"models",
"nervous",
"system",
"social",
"sciences",
"mathematical",
"models",
"neuroscience",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"mathematics",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"syntax",
"bioinformatics",
"animal",
"studies",
"behavior",
"neural",
"pathways",
"mathematical",
"and",
"statistical",
"techniques",
"insects",
"probability",
"theory",
"grammar",
"arthropoda",
"psychology",
"eukaryota",
"neuroanatomy",
"anatomy",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] | 2019 | Drosophila melanogaster grooming possesses syntax with distinct rules at different temporal scales |
Cooperation is ubiquitous across all levels of biological systems ranging from microbial communities to human societies . It , however , seemingly contradicts the evolutionary theory , since cooperators are exploited by free-riders and thus are disfavored by natural selection . Many studies based on evolutionary game theory have tried to solve the puzzle and figure out the reason why cooperation exists and how it emerges . Network reciprocity is one of the mechanisms to promote cooperation , where nodes refer to individuals and links refer to social relationships . The spatial arrangement of mutant individuals , which refers to the clustering of mutants , plays a key role in network reciprocity . Besides , many other mechanisms supporting cooperation suggest that the clustering of mutants plays an important role in the expansion of mutants . However , the clustering of mutants and the game dynamics are typically coupled . It is still unclear how the clustering of mutants alone alters the evolutionary dynamics . To this end , we employ a minimal model with frequency independent fitness on a circle . It disentangles the clustering of mutants from game dynamics . The distance between two mutants on the circle is adopted as a natural indicator for the clustering of mutants or assortment . We find that the assortment is an amplifier of the selection for the connected mutants compared with the separated ones . Nevertheless , as mutants are separated , the more dispersed mutants are , the greater the chance of invasion is . It gives rise to the non-monotonic effect of clustering , which is counterintuitive . On the other hand , we find that less assortative mutants speed up fixation . Our model shows that the clustering of mutants plays a non-trivial role in fixation , which has emerged even if the game interaction is absent .
Cooperation is ubiquitous in the natural world ranging from microbial communities to human societies . Yet , it is seemingly against evolutionary theory , since cooperators forgo their own interest to benefit others whereas defectors pay nothing to get the benefit . The past two decades have seen an intensive study on how cooperation evolves via natural selection [1–10] . One of the key mechanisms to promote cooperation is network reciprocity . It assumes that individuals only interact with their neighbors . Consequently , either reproduction or competition for survival happens locally , which is not true for evolutionary dynamics in well-mixed population [3–7] . For network reciprocity , a simple rule has been derived [11] that cooperation is favored provided the benefit-to-cost ratio exceeds the average number of neighbors per individual . It holds for the Death-birth ( DB ) process under weak selection limit . A key intermediate step to achieve this simple rule is that a cooperator has more cooperator neighbors than defector neighbors . Furthermore , the fewer neighbors a cooperator has , the more proportion of cooperator neighbors a cooperator has . In other words , few neighbors per individual lead to the clustering of the cooperators for evolutionary dynamics on a network . A cooperator surrounded by many cooperator neighbors obtains high payoff , which facilitates the fixation of cooperation . This simple rule also paves the way to solve social dilemmas including those modeled by multi-player games [12] . Therefore , the assortment of cooperators has been intensively employed to investigate the fixation probability and the fixation time for stochastic evolutionary game dynamics on a network [3 , 4 , 9 , 13–15] . Besides , other mechanisms promoting cooperation also result in the assortment of cooperators as a key intermediate step [2 , 16 , 17] , which is similar to the network reciprocity . Therefore , it would be necessary to investigate how the assortment alters the evolutionary outcome . For previous studies on the evolution of cooperation on a network [3 , 4 , 6 , 7] , both the game interaction and assortment are taken into account . Typically cooperation is modeled as a social dilemma via dyadic or multi-player games [18] . The assortment of cooperators follows as a result of evolutionary dynamics ( for an exception , see [19] ) . It is still far from clear how assortment alone changes the fate of evolution . Here , we disentangle the game dynamics and the spatial clustering . And we establish a minimal model to explore this issue . To this end , we only consider the frequency-independent cases , without any game interactions , to explore the role that the clustering plays alone . As a first step , we adopt a circle as the underlying population structure . Our study starts with two mutants . They have an initial distance denoting the number of wild-type individuals between them . We regard the distance as a measure of the spatial assortment . And we explore how the assortment of mutants alters the fixation probability and fixation time analytically .
We assume that there are N individuals with two strategies , A ( wild-type ) and B ( mutant ) . The corresponding fitnesses are fA and fB , respectively . The fitness is frequency-independent . In other words , it is solely determined by the focal individuals’s strategy , and has nothing to do with its neighbors’ . All the individuals are located on a ring , i . e . , every individual has exactly two neighbors . We consider the Death-birth ( DB ) process . For each round , an individual is randomly chosen to die . Its two neighbors compete to reproduce an offspring who adopts the same strategy as its parent . The chance of successful reproduction is proportional to the neighbors’ fitnesses ( see Fig 1 for illustration ) . w denotes the number of mutants , and Sw is a state . Then the DB process is described by a one-dimensional Markov chain . The Markov chain has two absorbing states ( S0 and S6 ) and the other states ( Si , where 1 ≤ i ≤ 5 ) are of one equivalence class . Denote Pa , b as the transition probability from state Sa to Sb , the Kolmogorov backward equation is written as πw=Pw , w+1πw+1+Pw , w−1πw−1+ ( 1−Pw , w+1−Pw , w−1 ) πw , withπ0=0andπN=1 , ( 1 ) where πw is the fixation probability starting from Sw . The fixation probability is then obtained [20]: πw=1+∑j=1w−1∏k=1jγk1+∑w=1N−1∏k=1jγk , ( 2 ) where γw=Pw , w−1Pw , w+1 . ( 3 ) Let r be the ratio of fitnesses between wild-type and mutant , i . e . , fAfB=r . It holds as follows: γw={r+12r , w=11r , w=2 , … , N−22r+1 , w=N−1 . ( 4 ) Taking Eq ( 4 ) into Eq ( 2 ) leads to the fixation probabilities . For the population size N = 6 , the fixation probability for two connected mutants is given: π2=r3 ( 1+3r ) 3+2r+2r2+2r3+3r4 . ( 5 ) Let τiA denote the conditional fixation time from state Si to SN , which refers to the mean time to absorption in state SN given the process starts in state Si and eventually reaches state SN . We have πiτiA=Pi , i−1πi−1 ( τi−1A+1 ) + ( 1−Pi , i−1−Pi , i+1 ) πi ( τiA+1 ) +Pi , i+1πi+1 ( τi+1A+1 ) , withπ0τ0A=0andτNA=0 . ( 6 ) Let us denote θi=πiτiA , then we arrive at a difference equation θi = Pi , i−1θi−1 + ( 1 − Pi , i−1 − Pi , i+1 ) θi + Pi , i+1θi+1 + 1 with boundary conditions θ0 = 0 and θN = 0 [8] . In particular , for θ0=π0τ0A , τ0A is infinitely large since it takes forever for the mutant to fixate if there is no mutant initially . On the other hand , π0 = 0 . We thus assume θ0 = 0 as in [8] . Solving the recursive equations [8 , 21] leads to τiA=τ1Aπ1πi∑k=1i−1∏m=1k−1γm−∑k=1i−1∑l=1k−11πiπlPl , l+1∏m=l+1kγm , withτ1=∑k=1N−1∑l=1kπlPl , l+1∏m=l+1kγm . ( 7 ) Taking N = 6 into the above equation , we obtain τ2A=3 ( 11+75r+132r2+140r3+109r4+45r5 ) ( 1+3r ) ( 3+2r+2r2+2r3+3r4 ) . ( 8 ) To explore the effect of the spatial clustering , we consider the process that there are two mutants with distance d in the beginning . That is to say , there are d connected wild-type individuals located between two mutants initially . In this section , we take N = 6 as an illustrative case . Note that six is the minimal size of a circle , in which there are two kinds of unconnected mutants . All the circles with population size below six have none or one such network configurations , as shown in Fig 2 . As illustrated in Fig 3 , the process gives rise to more states than that which starts with two connected mutants . Comparing with the previous process in Fig 1 , we divide all the states into two sets: the middle-state set S and the final-state set F . The middle states refer to all the states with two separated groups of mutants whereas the final states contain only one mutant group . Note that a group refers to connected individuals with the same strategy . Fig 3 shows four properties of the process: i ) All the middle states reach each other and belong to one equivalence class . ii ) The final states reach each other and belong to one equivalence class ( Fi , 1 ≤ i ≤ 5 ) and two absorbing states ( F0 and F6 ) . iii ) The middle states reach final states in finite time; however , the final states cannot reach any middle states . iv ) The middle states are transient ( sooner or later , they walk into one of the final states ) . These four features imply that the underlying Markov chain is not one-dimensional anymore , which leads to both computational and analytical challenges . The transition matrix P for the process in Fig 3 is listed as follows: ( 9 ) Denote Ψi as the fixation probability starting from state i ∈ {S , F} and ending up with state F6 . Based on the Markov property , the following holds: Ψi=∑j∈{S , F}Pi , jΨj , ∀i∈{S , F} . ( 10 ) It is equivalent to Ψ=PΨ , ( 11 ) with boundary conditions ΨF0=0 and ΨF6=1 ( subject to the property ii ) ) . We divide the states into two sets: the middle-state set S and the final-state set F . We denote Ψ= ( ΨSΨF ) . As the two crossing lines in Eq ( 9 ) illustrates , the one-step transition matrix P can be written as P=SFSF ( Q1Q20Q3 ) . ( 12 ) The transition probability from F to S is a zero matrix , which arises from property iii ) . In addition , we have that the sub-matrix Q2 is independent of mutant fitness r . In fact , the entries in Q2 implies the transition probability whose event is the collapse of two separate groups with the same strategy . Here , a group refers to connected individuals with the same strategy . Take the transition from S1 to F1 as an example , the transition occurs when a mutant is chosen to die with probability 26 . The chosen mutant has two wild-type neighbors . In this case , the chosen mutant will , with probability one , be replaced by a wild-type offspring . Thus , the transition probability from S1 to F1 is independent on the relative fitness of the mutant r . In general , this applies to any transition from the middle-state set S to final-state set F . Therefore , Q2 is independent of mutant fitness r . Similarly , Q1 and Ψ are dependent on r . Taking Eq ( 12 ) into Eq ( 11 ) , we obtain {ΨS=Q1ΨS+Q2ΨFΨF=Q3ΨF . ( 13 ) Note that the second equation ( ΨF = Q3ΨF ) is the same as Eq ( 1 ) . Thus we have ΨF = π . We now consider the first equation ( ΨS = Q1ΨS + Q2ΨF ) , which can be transferred to ( I − Q1 ) ΨS = Q2ΨF . We show that I − Q1 is invertible in the following . Since all the middle states are transient with respect to process property i ) and iv ) , we have ∑t=0∞Pij ( t ) <∞ , ∀i , j∈S [22] . This is equivalent to ( ∑t=0∞Q1t ) ij<∞ . ( 14 ) Let H=∑t=0∞Q1t , and we know H exists . Notice that H ( I−Q1 ) =∑t=0∞Q1t ( I−Q1 ) =∑t=0∞Q1t−∑t=1∞Q1t=I , ( 15 ) where I is the identity matrix with the same size as that of Q1 . This shows that H is the left-inverse of ( I − Q1 ) , and a similar argument shows that H is also the right-inverse of ( I − Q1 ) . We then acknowledge that H = ( I − Q1 ) −1 . Thus , ( I − Q1 ) is invertible . Thus , it holds ΨS= ( I−Q1 ) −1Q2ΨF . ( 16 ) In the process of Fig 3 , the fixation probabilities of states with two separated mutants are listed as follows: {ΨS1=r2 ( 28+167r+475r2+920r3+1036r4+585r5+117r6 ) ( 3+2r+2r2+2r3+3r4 ) ( 45+210r+322r2+210r3+45r4 ) ΨS2=r2 ( 39+179r+456r2+896r3+1041r4+597r5+120r6 ) ( 3+2r+2r2+2r3+3r4 ) ( 45+210r+322r2+210r3+45r4 ) . ( 17 ) We now investigate how long it takes for mutants to reach the state consisting of only mutants . Let τiA be the average conditional fixation time from state i ∈ {S , F} to F6 , given the population ends up with all mutants , i . e , F6 . TA is a vector of τiA , and it is denoted as TA= ( TSATFA ) , where TSA is the conditional fixation time to F6 for the middle states and TFA is that for the final states . For state i ∈ {S , F} , we have Ψi·τiA=∑j∈{S , F}Ψj·Pi , j· ( τjA+1 ) , withτF6A=0andΨF0·τF0A=0 . ( 18 ) The right side of Eq ( 18 ) contains all the cases that one-step further from S1 , weighted by the one-step transition probabilities . The symbol ∘ is the Hadamard product . For two matrices A = [aij] and B = [bij] with the same dimensions , we have A ∘ B = [aij ⋅ bij] . We then transfer Eq ( 18 ) to Ψ∘TA=P·[Ψ∘ ( TA+1 ) ] , ( 19 ) where 1 is a vector of value 1 with the same dimensions as TA . This is equivalent to Ψ∘TA=P· ( Ψ∘TA ) +P·Ψ . ( 20 ) And moving all elements with Ψ ∘ T to the left side , we have ( I−P ) · ( Ψ∘TA ) =P·Ψ , ( 21 ) where I is the identity matrix . Splitting the middle and final states , Eq ( 21 ) is written as ( I−Q1−Q20I−Q3 ) · ( ΨS∘TSAΨF∘TFA ) = ( Q1Q20Q3 ) · ( ΨSΨF ) . ( 22 ) We then obtain { ( I−Q1 ) · ( ΨS∘TSA ) −Q2· ( ΨF∘TFA ) =Q1·ΨS+Q2·ΨF0− ( I−Q3 ) · ( ΨF∘TFA ) =0+Q3·ΨF . ( 23 ) We have a solution for Eq ( 23 ) based on [8] . In particular , for j = 1 , … , 5 , we have τFjA=τF1AΨF1ΨFj∑k=1j−1∏m=1k−1γm−∑k=1j−1∑l=1k−1ΨFlΨFj1PFl , Fl+1∏m=l+1kγm . ( 24 ) We now look into the first equation in Eq ( 23 ) . Note that we have proved ( I − Q1 ) is invertible , thus we have TSA= ( I−Q1 ) −1 ( Q1·ΨS+Q2·ΨF+Q2· ( ΨF∘TFA ) ) ⊘ΨS , ( 25 ) where ⊘ is the Hadamard division operator . And Eq ( 25 ) is equivalent to TSA= ( I−Q1 ) −1[Q1· ( I−Q1 ) −1Q2ΨF+Q2·ΨF+Q2· ( ΨF∘TFA ) ]⊘[ ( I−Q1 ) −1Q2ΨF] . ( 26 ) We do not present the analytic expressions here due to the great complexity of the expression of τS1A and τS2A . We now address the DB process on a circle for arbitrary size . We denote the population size as N . A group refers to connected individuals with the same strategy . As Fig 4 illustrated , each state corresponds to a triplet ( x , a , b ) : x is the minimal distance of two mutant groups , a is the population size of the smaller mutant group and b is the population size of the larger mutant group . Note that the larger distance between two mutant groups equal to N − x − a − b . All the states for process for population size N are listed in Table 1 . We divide the states into the middle-state set S and the final-state set F . When the minimal distance x between two mutant groups is zero ( x = 0 ) , or when the number of one of the mutant groups is 0 ( a = 0 ) , we use F to replace the triplet expression Sx , a , b ( S0 , a , b = Fa+b and Sx , 0 , b = Fb ) . The middle states have two separate mutant groups whereas the final states only have one . From Table 1 , we find that the total number of states is ∑w=0N⌊w2⌋·⌊N−w2⌋+N+1 . ( 27 ) The total number of states is of O ( N2 ) , since ( ∑w=0Nw2·N−w2+N+1 ) is O ( N2 ) . Difficulty arises to calculate the fixation probabilities with the equation ΨP = Ψ . For every transition between the states in Table 1 , the state Sx , a , b stays where it is , or transits to a state where the mutant number is one greater or one less . Note that the mutant number equals to the sum of two mutant group sizes a + b . Take S1 , 1 , 1 for an example , it can transit to itself , S1 , 1 , 2 , S2 , 1 , 2 or F1 ( = S1 , 0 , 1 ) . The state transition only occurs when an individual at the border of a group is chosen to die . As the mutants and wild-type individuals have one or two groups respectively , there are at most 8 individuals on the border . That is to say , starting from any state , there are at most 8 transitions . If the population size N is large , the transition matrix is sparse . We list all the transition probabilities and boundary conditions in Table 2 . The process for arbitrary population size shares the same four properties as the process of population size N = 6: i ) The middle states reach any other middle states and belong to one equivalence class . Take the transition from S1 , 1 , 1 to S2 , 2 , 2 for an instance , there is a path as S1 , 1 , 1 → S1 , 1 , 2 → S2 , 1 , 1 → S2 , 1 , 2 → S2 , 2 , 2 . Going through this path , the transition number in Table 2 occurs in the order #5 → #11 → #5 → #4 . ii ) The final states contain three equivalence classes . One is F0 , one is FN , and all the rest give rise to the other equivalence class . iii ) The middle states reach final states in finite time , whereas the final states cannot reach any middle states . iv ) The middle states are transient states . With the four properties , the analysis from Eqs ( 11 ) to ( 16 ) still apply here . We obtain the fixation probabilities of mutants for the process for arbitrary population size by ΨS= ( I−Q1 ) −1Q2ΨF , ( 28 ) where Q1 and ΨF are dependent on the relative mutant fitness r , whereas Q2 is independent on r . Similarly , we obtain the conditional fixation time for mutants with arbitrary population size as TSA= ( I−Q1 ) −1[Q1· ( I−Q1 ) −1Q2ΨF+Q2·ΨF+Q2· ( ΨF∘TFA ) ]⊘[ ( I−Q1 ) −1Q2ΨF] . ( 29 ) In S2 Appendix , we develop an algorithm to numerically obtain Q1 and Q2 with a time complexity of O ( N2 ) and a space complexity of O ( N4 ) ( O ( N2 ) if sparse matrix method is employed ) . Combining with Eqs ( 28 ) and ( 29 ) , we have the fixation probabilities and conditional fixation times for mutants with arbitrary population size N . As the algorithm makes use of matrix multiplications and inversions , the time complexities to obtain the fixation probability and conditional fixation time are both of O ( N4 . 746 ) [23–25] . In particular , with Taylor’s expansion around r = 1 for Eq ( 28 ) , we have ( see S3 Appendix ) Ψ ( r ) =Ψ ( 1 ) +ddrΨ ( r ) |r=1 ( r−1 ) +12d2dr2Ψ ( r ) |r=1 ( r−1 ) 2+o ( ( r−1 ) 2 ) , ( 30 ) where ddrΨ ( r ) =[I−Q1 ( r ) ]−1ddrQ1 ( r ) [I−Q1 ( r ) ]−1Q2π ( r ) +[I−Q1 ( r ) ]−1Q2ddrπ ( r ) , ( 31 ) d2dr2Ψ ( r ) =2[I−Q1 ( r ) ]−1ddrQ1[I−Q1 ( r ) ]−1ddrQ1 ( r ) [I−Q1 ( r ) ]−1Q2π ( r ) +[I−Q1 ( r ) ]−1d2dr2Q1 ( r ) [I−Q1 ( r ) ]−1Q2π ( r ) +2[I−Q1 ( r ) ]−1ddrQ1 ( r ) [I−Q1 ( r ) ]−1Q2ddrπ ( r ) +[I−Q1 ( r ) ]−1Q2d2dr2π ( r ) . ( 32 ) The algorithm we developed also applies to calculate the derivatives of the fixation probabilities . The derivative of a matrix is defined as the matrix of derivatives of corresponding item . We obtain ddrQ1 ( r ) by turning values in Table 2 into their first-order derivatives and running through the algorithm in S2 Appendix . Following Eq ( 31 ) , we obtain the first-order derivatives of the fixation probabilities . Besides , the time complexity is the same order as that of the fixation probability . Matrix multiplications and additions are required but they do not increase the time complexity . The required space is doubled but it is still of complexity O ( N4 ) ( O ( N2 ) for adopting sparse matrices ) . Similarly , we find that the higher-order derivatives of the fixation probabilities require only the same-order or lower-order derivatives of Q1 . Thus , we obtain the second-order derivatives and higher-order ones by turning the values in Table 1 to their higher-order derivatives . The overall complexity stays the same .
We have already obtained the fixation probabilities of the mutants for a circle with population size N = 6 based on Eqs ( 5 ) and ( 17 ) . Expanding the equations around neutral selection , i . e . r = 1 , gives rise to {ΨF2 ( =π2 ) =26+712 ( r−1 ) −40288 ( r−1 ) 2+o ( ( r−1 ) 3 ) ΨS1=26+612 ( r−1 ) −49288 ( r−1 ) 2+o ( ( r−1 ) 3 ) ΨS2=26+612 ( r−1 ) −46288 ( r−1 ) 2+o ( ( r−1 ) 3 ) . ( 33 ) Note that F2 refers to the state of two connected mutants , while S1 and S2 refer to the states where two mutants are in distance of 1 and 2 , respectively . Base on Eq ( 33 ) , we find that on the circle with population size 6: i ) Under neutral selection , i . e . r = 1 , the fixation probabilities are only determined by the number of the mutants . It has nothing to do with the distance of mutants . ii ) The first-order derivatives of fixation probabilities at r = 1 for separated mutants are equal ( ddrΨS1=ddrΨS2=612 ) . They are greater than 0 , but are smaller than that of the connected mutants ( ddrΨF2=712 ) . This indicates a slower change for the separated mutants than the connected mutants in fixation probability , as Fig 5 shows . In particular , if r < 1 , i . e . , the mutants are at a disadvantage , the fully connected mutants weaken the fixation probability . The fully connected mutants greatly promote the invasion when they are at an advantage ( i . e . , r > 1 ) . iii ) The second-order derivative of the fixation probability at r = 1 for d = 1 is smaller than that for d = 2 . Here , the second-order derivative of the fixation probability is two times of the second-order coefficient of the Taylor series . Thus , the closer the two mutants are , the less likely the invasion probability is under strong selection . Consequecntly , the rank of the invasion chances is determined solely by the clustering factor , i . e . , the distance of two mutants , as long as mutants are not fully connected . Using the developed algorithm , we generalize the above results on a small circle with population size 6 to a circle with large size . We investigate the fixation probabilities for population size 25 in Fig 6 . Not all the distances are plotted with only d = 1 , 2 , 3 , 11 shown in Fig 6 . This is because there are so many to show , and they do not lead to novel insights . In addition , we list fixation probabilities and their derivatives at neutral selection numerically in Table 3 for population sizes N = 6 , 25 , 100 respectively . We have found similar properties as that in the small one ( Fig 6 ) : i ) The fixation probabilities are proportional to the number of mutants at neutral selection , i . e . , r = 1 . ii ) The first-order derivatives at r = 1 for separated mutants ( d > 0 ) are the same . The first-order derivatives at r = 1 for the connected mutants are greater than that of the separated mutants . For N = 6 , 25 , 100 , we find that the first-order derivatives of fixation probabilities at neutral selection can be summarized as {ddrΨF2 ( =ΨS0 , 1 , 1 ) |r=1=2N−52NddrΨSd , 1 , 1|r=1=2N−62N , d=1 , 2 , . . . , ⌊N−22⌋ . ( 34 ) This can apply ∀N ≥ 6 , but the proof is still an open issue . iii ) For the separated mutants , the second-order derivative of the fixation probabilities at r = 1 increases as the distance d grows . Thus , the rank of the invasion chances is determined only by the assortment factor , provided that the mutants are not initially connected . In this case , the greater the distance two mutants are , the greater the fixation probabilities are . Note that from N = 6 to N = 25 , the second-order derivatives of the fixation probabilities at neutral selection increase from negative to positive . It implies that the fixation probability as a function of the selection intensity r turns from convex to concave , as population size increases . Taking N = 6 into Eq ( 26 ) , we have the analytical results of the conditional fixation times τS1A and τS2A . Due to the complexity of the expressions , we do not present them here . Expanding Eq ( 8 ) , τS1A and τS2A around neutral selection , i . e . r = 1 , results in {τF2A=83226+312416 ( r−1 ) −786504992 ( r−1 ) 2+o ( ( r−1 ) 3 ) τS1A=76726+1125416 ( r−1 ) −854094992 ( r−1 ) 2+o ( ( r−1 ) 3 ) τS2A=73926+1422416 ( r−1 ) −880014992 ( r−1 ) 2+o ( ( r−1 ) 3 ) . ( 35 ) Fig 7 presents the analytical predictions , which are validated by the simulations: i ) At neutral selection r = 1 , the times for mutant fixation differ for different mutant distances , though the fixation probabilities are the same . The greater the distance two mutants is , the shorter it takes for the mutant to fixate . An intuitive explanation is that as the distance between two mutants grows , each mutant becomes more independent as a source for strategy spreading . This is similar to infection sources in epidemiology . More infection sources speed up mutant fixation . For instance , when two mutants are connected ( d = 0 ) , there are initially only 2 wild-type individuals that can be updated . On the contrary , when d = 2 , there are 4 wild-type individuals that can be updated . ii ) When the mutants are at an advantage ( r > 1 ) , the conditional fixation time and the fixation probability are nontrivial . On the one hand , for each mutant distance , the mutant conditional fixation time grows at first and decrease as the mutant fitness r grows . On the other hand , given a constant mutant fitness r , when the distance between two mutants d becomes greater , the fixation time shrinks whereas the fixation probability changes non-monotonically ( it decreases to the least when d = 1 and grows when d > 1 ) . iii ) When the mutants are disadvantageous ( r < 1 ) , mutants have a better chance to fixate and also fixate faster as the distance between two mutants grows . In general , the fixation probability and the conditional fixation time for mutants do not have the same tendency as mutants are getting clustered [26] . And the rank of times for mutant fixation is monotonically determined by the clustering factor ( i . e . , the distance between mutants ) . Fig 8 presents the simulation results of the conditional fixation time of mutants and the numerical curve ( calculated by our algorithm ) for population size N = 25 . We observe that it agrees perfectly with the theoretical predictions .
Cooperation plays a key role in all levels of biological systems . Network reciprocity , as one of the mechanisms to promote cooperation , has attracted considerable interests . Network reciprocity results in an assortment between individuals using the same strategy [9 , 13 , 14 , 27] . Besides network reciprocity , the tag-based dynamics also yields more frequent interactions within groups equipping with the same tag [2 , 16 , 17] . This in-group bias is created via the tag . In this case , individuals who donate to others with the same tag protect the cooperators from being exploited [2 , 17] . The clustering individuals with the same tag lead to a high donation level . Again , the clustering here shows itself as an important intermediate step to facilitate cooperation . The assortment of mutants can either promote [11] or inhibit cooperation [4] . For example , the intensive interaction between individuals using the same strategy can be beneficial for cooperation in the Prisoners’ Dilemma , yet can be destructive for cooperation in the Snowdrift Games [28] . The nontrivial role the assortment plays can also be suggested from the previous studies: Fu et al . [29] have compared invasions of the Snowdrift Game with that of the Prisoners’ Dilemma on a lattice . As the cost-to-benefit ratio grows , the mutants tend to emerge as few large compact clusters in the Prisoner’s Dilemma whereas the mutants evolve to many dispersal small clusters in the Snowdrift Games . All these previous studies are based on game interactions . It is not clear how the assortment alone affects evolutionary dynamics . Inspired by these , we try to disentangle the spatial assortment of mutants from the game interaction . We implemented a minimal model via adopting a circle as the spatial structure . We assume that the fitness is frequency-independent . And the assortment of two mutants is easily measured by the minimum number of wild-type individuals in between . In real biological systems , there can be three reasons for mutants to be spatially separated: i ) Independent mutations . The mutant individuals who are not spatially adjacent arise via independent mutations; ii ) Migration . One of the mutant individuals , originally adjacent to the others , migrate to another place and settle there; iii ) The mutants are separated due to sudden environmental changes . As an illustrative case , we study the process with population size N = 6 . It is the minimum size of a circle , where two mutants can be of three different distances ( Fig 2 ) . We adopt the Death-birth process on a network . The analytical results show that initially fully connected mutants enhance the group survival when they are at an advantage ( r > 1 ) , whereas inhibit survival when mutants are at a disadvantage ( r < 1 ) . The simulation results are found to be in perfect agreement with the analytical ones ( Fig 5 ) . In other words , the spatial assortment of mutants is an amplifier of natural selection for the connected mutants compared with the separated mutants . However , as long as two mutants are separated , the relative mutant fitness r does not determine the rank of the probability of successful invasion , the distance between two mutants does . Denoting d as the initial distance between two mutants , the fixation probability falls to the smallest value when d = 1 , and it grows as d becomes greater . That is to say , as long as the mutants are separated , the further they initially are , the greater the invasion chance is . It is true for both advantageous and disadvantageous mutants . Our results show that the effect of spatial clustering on fixation is non-trivial , even when the game interaction is absent . The fixation probability for the separated mutants cannot be obtained as easily as obtaining that of the connected mutants . On the one hand , the separated mutants introduce many additional states; on the other hand , the resulting Markov chain is not one-dimensional anymore . We further categorize the transient states into two classes . And we make use of the fixation probabilities for the connected mutants to obtain the fixation probabilities for the separated mutants . In addition , we have developed an efficient algorithm to estimate the fixation probability for the separated mutants in arbitrary population size . In general , the algorithm consists of three steps: 1 ) Listing all the states of the Markov chain in order , 2 ) Listing all the transition probabilities in S2 Appendix , 3 ) calculate the derivatives in Eqs ( 31 ) and ( 32 ) based on S3 Appendix . The time complexity is of O ( N4 . 746 ) . The space complexity is of O ( N4 ) , and O ( N2 ) if sparse matrix methods are adopted [23–25] . We evaluated the processes for population sizes of 25 and 100 . All the above-mentioned results still apply as in the small circle with size 6 . Thus , we conjecture that the main results apply to any population size . Yet the strict proof is still an open issue . In the work of Ohtsuki et al . [11] , the pair approximation is adopted . Under weak selection limit , it is only the initial frequency of mutants that is key to the fixation probabilities . The fixation probability has nothing to do with the assortment parameter . There , the assortment parameter is given by qA|A − qB|A [27 , 30 , 31] , which refers to the difference between the number of neighbors using the same strategy and that of the ones using the other strategy . Our results show a different picture . Let us take a ring consisting of 25 individuals with 2 mutants as an example . When the distance between two mutants are equal or greater than two ( 2 ≤ d ≤ 12 ) , all the cases share the same probability of finding a mutant next to a wild-type individual ( qB|A=423 ) or finding a wild-type individual next to a wild-type individual ( qA|A = 1 ) . By pair approximation , the fixation probability is the same for all the mentioned initial population configurations , since the number of mutants is two , and even the assortment factor qA|A−qB|A=1923 is the same . However , our analytical results show that the fixation probabilities can be in a large difference when the selection intensity is strong , verified by simulations . Furthermore , we show that the difference occurs at the second-order derivative of the selection intensity . This suggests that the number of initial mutants alone cannot determine the fixation probability . Therefore , the pair approximation is not sufficient to portray the spatial clustering accurately , provided the selection intensity is strong . We investigate the cases with Death-birth process . The DB process contains two steps: An individual is randomly chosen to die and its nearest neighbors compete to reproduce an identical offspring . The assumption behind the DB process is that the death rate is equal for all the individuals and the selection happens in the stage of reproduction . The competition for reproduction is local . The Birth-death ( BD ) process is different from the DB process . It contains two steps as well: An individual is chosen to produce an identical offspring with the possibility proportional to its fitness across the entire population , and the offspring replaces a neighbor of its parent randomly . The assumption here is that all the individuals compete for reproduction , thus the selection is global . The death rate is equal for all the neighbors . Noteworthily , for the BD process , the isothermal theorem [32] shows that the fixation probability of mutants with BD process is identical to that in well-mixed populations on all the isothermal graphs . A circle is an isothermal graph . It implies that the assortment of the mutants does not play any role in fixation probability , provided the number of mutants is the same for a circle with BD process . However , our results with the DB process depicts a different picture . Consequently , the details of update rule could dramatically alter the evolutionary dynamics on networks , even if the game interaction is not at work . Intuitively , the assortment plays a role if the competition for reproduction is local , but is not at work if the competition for reproduction is global . In fact , it has been shown that different evolutionary rules can alter the evolutionary outcome not only in the networked population [11] but also in a simple well-mixed population [33] . Our results also echo the recent studies that the DB process does not conform to the isothermal theorem [10 , 34] . To sum up , our results reveal counterintuitive but fundamental effects of spatial clustering on the evolutionary dynamics . In particular , the clustering plays its role without the involvement of games . It is not hard to imagine the great complexity arises when games are involved or when complex graphs are introduced . This deserves further studies . In addition , our model can be used as a reference case to better understand how the clustering of mutants favor or disfavor cooperation . Furthermore , it is a natural association that our conclusion calls for biological experiments , such as the microbial experiments , to verify the effect of spatial assortment on evolutionary dynamics . | Evolutionary dynamics on networks are key for biological and social evolution . Typically , the clustering mutants on networks can dramatically alter the direction of selection . Previous studies on the assortment of mutants assume that individuals interact in a frequency-dependent way . It is hard to tell how assortment alone alters the evolutionary fate . We establish a minimal network model to disentangle the assortment from the game interaction . We find that for weak selection limit , the assortment of mutants plays little role in fixation probability . For strong selection limit , connected mutants , i . e . , the maximum assortment , are best for fixation . When the mutants are separated by only one wild-type individual , it is worse off than that separated by more than one wild-type individual in fixation probability . Our results show the nontrivial yet fundamental effect of the clustering on fixation . Noteworthily , it has already arisen , even if the game interaction is absent . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] | [
"recreation",
"organismal",
"evolution",
"markov",
"models",
"statistics",
"applied",
"mathematics",
"microbiology",
"social",
"sciences",
"simulation",
"and",
"modeling",
"algorithms",
"mathematics",
"network",
"analysis",
"microbial",
"evolution",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"games",
"behavior",
"probability",
"theory",
"population",
"metrics",
"psychology",
"network",
"reciprocity",
"population",
"size",
"natural",
"selection",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"processes",
"markov",
"processes"
] | 2019 | Close spatial arrangement of mutants favors and disfavors fixation |
Frequency tuning and phase-locking are two fundamental properties generated in the cochlea , enabling but also limiting the coding of sounds by the auditory nerve ( AN ) . In humans , these limits are unknown , but high resolution has been postulated for both properties . Electrophysiological recordings from the AN of normal-hearing volunteers indicate that human frequency tuning , but not phase-locking , exceeds the resolution observed in animal models .
The cochlea decomposes sound into bands of frequencies and encodes the temporal waveform in these bands , generating frequency tuning and phase-locking in the auditory nerve ( AN ) . The relative roles of these two processes in human perception have long been debated [1 , 2] and would be clarified by knowing their limits . For example , studies in animals show that the average firing rate of AN fibers codes the spectral envelope of human vowels , but this code is problematic at high sound intensities . In contrast , a code based on phase-locking of the sound’s waveform ( its “fine-structure , ” i . e . , the fast fluctuations in instantaneous pressure ) is adequate at all intensities but does not extend above a few kilohertz . The difficulties of both coding schemes in explaining human perceptual ranges may reflect physiological differences in resolution between animal models and humans , in whom single fibers cannot be studied . The problem of rate coding at high sound levels may reflect broader spectral filtering in animal models [3] . This is supported by indirect estimates that report exceptionally sharp frequency tuning in humans , using behavioral estimates or otoacoustic emissions in subjects with normal hearing [4 , 5] and mass potentials in patients [6 , 7] . However , this conclusion is disputed [8–10] . The upper frequency limit of phase-locking is species dependent [11] but is unknown in human . Some perceptual abilities suggest use of temporal cues at 10 kHz or higher [12–16] , but binaural sensitivity implies an upper limit barely above 1 kHz [17 , 18] . In summary , the present evidence regarding the limits of frequency tuning and phase-locking is conflicting . Knowledge of these limits is also important to understand and treat human hearing impairment [12 , 13 , 19] . We modified a clinical electrophysiological method [6 , 7 , 20 , 21] to study the AN in normal-hearing humans and macaque monkeys . An electrode is inserted through the eardrum to record potentials from the cochlear bony capsule . Combining a closed acoustic system calibrated in situ , stable trans-tympanic electrode placement under visual control , and validated stimulus and analysis paradigms [22 , 23] , we studied the AN over several hours . Frequency tuning was obtained using pure tones to probe the imprint of a spectrally manipulated preceding notched-noise masker on the compound action potential ( CAP; the summed response of AN fibers at the onset of the probe tone ) . Neural phase-locking was assessed with a paradigm separating the nonmaskable cochlear microphonic ( CM ) generated by hair cells , from the AN neurophonic . We achieved our aim of measuring both frequency tuning and the limit of phase-locking in humans and macaque monkeys and found that humans are unusual in the sharpness of frequency tuning but do not excel in the upper frequency limit of phase-locking .
The sharpness of frequency tuning obtained with the notched-noise forward-masking ( NNFM ) paradigm is shown in Fig 1 as a quality factor ( Q10 ) . Both human ( Fig 1a ) and monkey ( Fig 1b ) show a monotonic increase with probe frequency , consistent with other species [22 , 24] and with the vast literature on single AN fibers , but Q10 values in humans are significantly higher than in other species ( Fig 1c ) when compared over the same frequency range ( average factors: approximately 1 . 6 times cat and chinchilla and 1 . 3 times monkey ) . CAP-based Q10 values differ from values in single AN fibers [22] , which are the ultimate reference but cannot be studied in humans . However , availability of the 2 sets of data ( CAP and single unit ) in animals allows calculation of conversion functions based on the ratio between single-fiber Q10 and CAP-Q10 as a function of frequency . Applying the average of the conversion functions for cat , chinchilla , and macaque ( S5a–S5c Fig ) to measured human CAP-Q10 values , we predict human single-fiber Q10 to be slightly above those for macaque monkey [24] ( red versus blue solid line , Fig 2 ) . Using only the monkey conversion function , which is arguably the most relevant , the predicted human single-fiber trend is even higher ( red dotted-dashed line , Fig 2 ) and is remarkably consistent with assessments using nonelectrophysiological techniques [4 , 5] ( Fig 2 , green lines ) . Both predicted trendlines are higher than Q10 values reported for smaller , nonprimate animal models ( Fig 2 , red lines versus shaded area ) . Phase-locking in human and monkey showed the band-pass characteristic previously observed in similar measurements of the cat [23] . Maximal absolute amplitudes are only 4 dB smaller in human than in macaque; both are smaller than in cat [23] . The center frequency at the maximal absolute amplitude and the steep upper-frequency slope were lower in human ( at approximately 0 . 7 and 3 kHz ) than in monkey ( at approximately 1 and 4 kHz ) ( Fig 3a ) . At face value , the data suggest that the upper phase-locking limit is lowest in human . However , several factors affect the absolute amplitude of the measured signal so that there is an unknown vertical offset between data for different species . For example , the recordings in cat were taken with a ball electrode on the round window , while in humans and monkeys , a needle was placed on the cochlear bony capsule , which provided a much smaller signal . In Fig 3b , single-fiber data are used to anchor cat and monkey data ( S9 Fig ) while keeping the relative position of monkey and human data . The frequencies ( kilohertz ) at which the trendlines cross the abscissa are 4 . 7 ( cat ) , 4 . 1 ( monkey ) , and 3 . 3 ( human ) . Alternatively , the data for humans can be normalized to the maximum observed in cat ( Fig 3b , dotted line ) —even then , there is no suggestion of a higher limit of phase-locking in human than in cat .
We obtained the first electrophysiological recordings , to our knowledge , of cochlear potentials , which address both frequency tuning and temporal coding in humans with normal hearing . It is generally agreed that both processes have critical roles in human auditory perception , but there is considerable controversy regarding their relative roles , as well as regarding their resolution when compared to animal models . Impaired frequency selectivity and phase-locking have both been proposed as main causes for human hearing impairment [12 , 13 , 19] . The importance and presence of frequency tuning in humans is not under discussion , but 2 early studies that concluded that human frequency tuning is exceptionally sharp [4–6] were subsequently contradicted by different data or analyses [8–10] . Our data , which are electrophysiological and fundamentally different in nature from the previous estimates based on cochlear emissions and behavior [4 , 5] , are strikingly in line with these earlier estimates ( Fig 2 ) . An earlier study using a recording technique similar to ours [6] shows 2 CAP-Q10 values obtained with a tonal forward-masking paradigm at 8 kHz from 2 subjects with nominally normal hearing: these values ( 6 . 2 and 8 . 2 ) are reasonably in line with ( somewhat lower than ) the trend of our measurements extending to 6 kHz ( Fig 1 ) . From these measurements , the authors propose that human frequency tuning is sharper than in guinea pig and chinchilla but is rather similar to that in cat , which is not what we find when the same CAP-Q10 measurements are obtained in these different species ( Figs 1 and 2 and [22] ) . Recent behavioral data suggest that frequency tuning in monkeys is not as sharp as in humans [25 , 26] , consistent with our physiological data ( Fig 1 ) . The situation is somewhat different for coding of fine-structure in humans , for which the discussion has been entirely based on behavioral research , and no attempt has been made to obtain direct measurements in humans . It is undisputed that coding of sound fine-structure is a prerequisite for binaural temporal sensitivity at low frequencies , with an abrupt upper limit at approximately 1 . 3 kHz [27 , 28] . Such coding has been proposed to be important for other auditory attributes as well , at frequencies as high as 10 kHz or more [12–16] , but this is debated [17 , 29] . Using a validated technique [23] to extract neural phase-locking from the potentials measured near the cochlea , we find a reduced upper limit of phase-locking in monkey relative to cat , and in human relative to monkey . The consistency of these limits with those obtained in studies of single AN fibers ( for cat and monkey ) argue that phase-locking in human is limited to lower—rather than higher—frequencies than in commonly used laboratory species . Our findings suggest a reappraisal of the fundamental debate that has been ongoing in hearing science for more than a century , regarding the importance of temporal versus “place” coding . This debate has taken various forms but in the past decades has centered on different codes available in the AN . As one example , in studies of the coding of human speech sounds by the firing rate of single AN fibers in small laboratory animals , frequency selectivity and dynamic range were not sufficient to code spectral features over the behaviorally relevant range [2 , 3 , 30] . On the other hand , phase-locking can account for the extremely wide perceptual intensity range but is limited in the upper frequency to which such coding is present , and it remains unclear how that temporal information can be extracted by the central nervous system . Our results suggest that the limits faced by models of “place coding” are less severe , and those by models of “temporal coding” more severe , than was thought based on data obtained from the small animal species used in neurophysiological experiments . For place coding they are less severe because the place map in the human cochlea is expected to be more fine-grained than in the experimental species studied [1 , 3] . For temporal coding the phase-locking limit is more severe because the fine-structure of sounds will not be coded up to the high frequencies at which it is , e . g . , in the cat ( about 5 kHz ) [31] . Of course , humans differ from the other species studied along many dimensions , and it is at this point unclear how unique sharp frequency tuning is among mammals . A common , simple reasoning is that a small number of octaves “packaged” into a long cochlea will result in sharper frequency tuning than a large number of octaves subserved by a short cochlea [3 , 6 , 32–34] . Implicitly , this reasoning assumes that frequency tuning is limited by an absolute distance on the basilar membrane , which is similar across species . Measurements of cochlear dimensions in skulls of fossil and extant mammals [35] suggest that the cochlea of modern humans is “hypertrophied” relative to expectations on body size , so perhaps there is something special about the human use of hearing which drove sharper frequency selectivity . On the other hand , comparative studies suggest that larger animals have sharper tuning [32 , 33] , so possibly human tuning is only remarkable in sharpness when compared to the ( small ) species studied experimentally . Independent of this issue , the sharper tuning observed in humans relative to species used in physiological studies indicates that inferences toward human perception have to be made cautiously , particularly when spectral versus temporal schemes are considered . Data in animals were obtained under general anesthesia , whereas in humans only a local anesthetic was used: could our higher human Q-values be caused by this methodological difference ? Effects of activation of the middle ear reflex can be excluded because we were careful not to evoke this reflex . First , the reflex intensity threshold was measured in all subjects ( using tympanometry , see Materials and methods ) . Second , activation of the middle ear reflex can be monitored during the recordings because muscle action potentials strongly contaminate the neural recordings . We are also quite confident that efferents of the medial olivo-cochlear ( MOC ) system did not contribute to the sharper tuning observed in humans because we explicitly looked for efferent effects ( on CAP or neurophonic amplitude ) in separate experiments and found them to be very small ( a few decibels ) , and biased towards low frequencies [36] . The MOC reflex is functional in both anesthetized and decerebrated cats [37] , and no difference in CAP tuning was found in awake versus anesthetized guinea pigs [6] . Finally , the expectation from animal work is that efferent effects would cause a reduction in sharpness of tuning , so that if only present in awake humans , they would have tended to make the difference between Q-values in humans and anesthetized animals smaller rather than larger . There are marked difference in length of the ANs of the species studied ( factor of approximately 4 between cat and human ) : could the differences that we measured in the upper-frequency limit of phase-locking reflect spatial integration of the propagating action potentials along the AN ? Although this issue can only be directly addressed by recordings from individual nerve fibers , there are several arguments against such spatial integration . It would cause significant low-pass filtering of the CAP in human , while we find that the initial negative waveform in humans and cats is very similar ( e . g . , S2 Fig ) . Also , studies of mass potentials [38–40] and of unit contributions of single nerve fibers [41 , 42] suggest that these potentials reflect a potential difference generated over a restricted segment of the AN rather than spatially distributed generators . In summary , we provide electrophysiological evidence that our species excels in sharpness of frequency tuning but not in temporal coding of fine-structure . This dual result calls for a reappraisal of coding schemes based on average firing rate , e . g . , for the coding of pitch and speech . The plausibility of such schemes relative to temporal schemes may have been unduly dismissed based on the more limited resolution of place-rate coding in experimental animals on the one hand , and unrealistic assumptions regarding the extent of temporal coding in humans on the other hand .
Human experiments were carried out in accordance with the recommendations of good clinical practice ( ICH/GCP ) and were approved by the Medical Ethics Committee of the University of Leuven . All subjects gave written informed consent in accordance with the Declaration of Helsinki . Human volunteers were recruited on campus with advertisements . A total of 19 subjects ( 15 female , 4 male ) participated in the experiments . Participants were between 20 and 35 years old and received a financial compensation . Frequency tuning data were based on recordings from 9 subjects; for neural phase-locking , data of 7 subjects were used . In 4 subjects , both types of data were recorded . In the remaining 7 subjects , no useable data were obtained , for various reasons . In 3 subjects , the signal-to-noise ratio ( SNR ) was too poor ( in 1 restless person due to excessive muscle artifacts and in the others for unknown reasons ) . In 2 subjects , the needle could not be placed at the desired location because of a narrow and heavily curved ear canal . In 2 subjects , no measurements could be started due to practical issues . Animal procedures were approved by the Animal Ethics Committee of the University of Leuven . Recordings in Monkey were obtained from 1 ear in 4 rhesus monkeys ( Macaca mulatta ) , which were also involved in chronic visual experiments ( an 8 . 9-kg adult male , a 4 . 8-kg juvenile male , and 2 juvenile females of 6 . 3 kg and 4 . 7 kg; ages were between 4 and 7 years ) . Prior to the experiments , dissections on formaline-preserved temporal bones were performed to study the best trajectory and practice the placement of the needle electrode . The day before—or morning of—the session , the hearing of the volunteers was screened , including an inquiry for hearing problems , a pure tone audiogram ( thresholds <20 dB nHL , 125 Hz–8 kHz ) , tympanometry to assess middle ear function , and an otoscopic examination by an otolaryngologist . Subjects were requested to avoid exposure to loud sounds in the days preceding the experimental session . For monkey , the tympanic membrane was otoscopically checked after induction of the anesthesia . Human subjects were unsedated during the experiment . Before insertion of the needle electrode , the tympanic membrane and ear canal were locally anesthetized with Bonain’s solution ( equal volumes of cocaine hydrochloride , phenol , and menthol; aspirated after 30 minutes ) . Subjects usually had a short-lasting and vague sensation of touch during insertion of the electrode , which quickly disappeared . Recording in monkey was similar to that in human , with the main difference being the presence of general anesthesia . Induction was done with a mixture of ketamine ( 3 mg/kg ) and medetomidine ( intramuscular , 0 . 050 mg/kg ) . The same mixture was administered intravenously for maintenance through a venous cannula inserted for administration of lactated Ringer’s solution . The duration of the total experimental session , including the placement of the electrode , was between 4 and 6 hours . After the experiments , atipamezole ( intramuscular , 0 . 2 mg/kg ) was administered to reverse the sedative effect of medetomidine; after awakening of the animal , it was observed until it was freely moving about . To minimize electrical and acoustical interference , all experiments were conducted in a double-walled soundproofed and faradized booth ( Industrial Acoustics Company , Niederkrüchten , Germany ) . Before the experiment , human subjects chose a comfortable supine position on a bed and were asked to remain still during the trans-tympanic insertion of the needle electrode ( TECA; sterile monopolar disposable , 75 mm × 26 G , 902-DMG75-TP ) and the actual recordings . While in the sound booth , subjects and experimentalists were grounded to the booth via an antistatic wrist strap . Anesthetized monkeys were positioned on a heating pad with their heads restricted in a stereotactic frame and turned for ease of needle insertion . Core body temperature was maintained using a feedback-controlled homeothermic system ( Harvard Apparatus , Model 50–7129 ) . Eyes were coated with a thin layer of ophthalmic ointment ( Pfizer , Terramycine ) to prevent desiccation . For every human subject , a custom silicone earmold ( Dentsply , Aquasil Ultra XLV regular ) was made for acoustical reproducibility throughout the procedure and to preserve low frequency performance of the earphone speaker ( Etymotic , ER-2 or ER-1 ) . The earmold contained 2 casted openings for different manipulations , such as needle insertion , visualization , acoustic stimulation , and calibration . During most actions ( e . g . , placing of the needle electrode through one of the earmold’s openings ) , the ear canal and tympanic membrane were visualized by a rigid endoscope with camera ( R . WOLF , 8654 . 402 25 degree PANOVIEW; ILO electronic GmbH , XE50-eco X-TFT-USB ) through the other available opening of the earmold . In order to maintain the position of the needle electrode relative to the unrestricted head in human , a custom frame—consisting of a ring that was centered above the external ear and fastened around the subject’s head with Velcro straps—was used . On this ring , a needle holder allowed stable support of the needle electrode under slight tension in order to maintain good electrical contact . In monkey , the recording needle was secured by a mechanical micro-manipulator mounted on the stereotactic frame . The placement of the needle electrode was performed while visualizing the ear canal and tympanic membrane with a surgical microscope ( ZEISS , OPMI pico ) . Earmolds were made in situ , after placement of the needle electrode , with ear impression compound ( Microsonic ) . The calibration of the acoustic system ( ear canal and earmold ) was performed in situ with a closed-loop system using a tube earphone speaker ( Etymotic Research , ER1 or ER2 ) and a microphone ( Etymotic Research , ER-7C ) with a silicon probe close to the tympanic membrane . In humans , the calibration was done before placement of the needle electrode . In monkeys , calibration was performed after placement of the needle electrode with the silicon probe tube embedded in the earmold . Sound was delivered through one of the openings of the earmold via a plastic T-piece , which allowed access for the endoscope . During calibration and recording , all openings were sealed airtight except for a tiny opening in the plastic T-piece that prevented static pressure build-up . A trans-tympanic procedure was developed , extensively tested , and practiced on more than 20 fresh human cadavers in the university hospital . The sterile needle electrode was inserted by an ENT surgeon through one of the openings of the earmold that contained a short sterile plastic tube ( length <1 cm; diameter 2 mm ) . The needle electrode was placed , trans-tympanically ( 3rd quadrant ) , on the cochlear promontory or in the niche of the round window . The experimental session was terminated within 4 hours or when the subject expressed the desire to stop . The needle electrode was then pulled back , and the earmold was removed . The session was concluded with an otomicroscopic examination . In no cases was there an eardrum perforation larger than expected from the needle’s diameter ( 0 . 46 mm ) . Subjects were requested to keep the ear dry for 10 days following the recording session . An otolaryngologist was available during the weeks after the experiment to address any worries or for an additional checkup . Stimuli were generated with custom software and a digital sound system ( Tucker-Davis Technologies , system 2 , sample rate: 125 kHz/channel ) consisting of electromagnetically shielded earphone speaker ( Etymotic Research , ER-1 or ER-2 ) , a headphone driver ( HB7 ) , a digitally controlled analog attenuator ( PA5 ) , and a digital-to-analog converter ( PD1 ) . Acoustically evoked cochlear mass potentials were recorded using a low-noise differential preamplifier ( Stanford Research Systems , SR560 ) , as described in our previous publication [36] . The signal input was connected to the trans-tympanic needle electrode , the reference input was connected to an earlobe clamp coated with conductive gel , and the ground input was connected to a standard disposable surface electrode placed at the mastoid , also coated with conductive gel . All contacts were made on the side ipsilateral to the recording . The battery-operated preamplifier was galvanically isolated ( A-M systems , Analog stimulus isolator Model 2200 ) from the mains-powered equipment . Before the signal was recorded ( TDT , RX8 , approximately 100 kHz/channel , maximum SNR 96 dB ) , stored , and analyzed ( The Mathworks , Matlab ) , the signal was further amplified ( DAGAN , BVC-700A ) to a total gain of × 100 k and band-pass filtered ( 30 Hz–30 kHz; cut-off slopes 12 dB/octave ) . During the sessions , the most relevant signals were visualized on an oscilloscope ( LeCroy , WaveSurfer 24Xs ) and monitored with a loudspeaker outside the experimental booth . Recordings were averaged off-line over multiple repetitions ( between 128 and 1 , 024 , depending on background noise level ) to increase SNR . CAP responses were obtained by summing responses with alternating stimulus polarity and were additionally de-noised with a band-pass filter in range of the spectrum of the CAP . CAP amplitudes were measured between the first negative trough ( N1 ) and first positive peak ( P1 ) , or if P1 was not clearly defined , between N1 and the second positive peak ( P2 ) ; otherwise , they were measured between N1 and the positive maximum ( S2 Fig ) . CAPs reflect activity of many AN fibers [41 , 43] but are not frequency selective . To assess frequency selectivity , we used a modified NNFM paradigm [4 , 44] to extract masking tuning curves ( MTCs ) . Briefly , this involves measuring the CAP to a probe signal that is a short pure tone , fixed in level and frequency . The probe level is fixed at the SPL that results in a SNR of 18 dB , when the probe is given by itself . The probe tone is then preceded by a noise ( forward ) masker , which results in a reduction of the CAP to the tone . First , a broadband noise masker is used whose SPL is adjusted so that that a CAP suppression of 33% is obtained . This masker is then increased 10 dB in level ( causing more masking ) ; a spectral notch is introduced centered at the frequency of the probe tone , and the notch width is then varied to search for the width restoring a CAP suppression of 33% . We assess neural phase-locking using a phase-locked neural component in the electrical mass potential recorded in the middle ear . Previously , we developed a method based on forward masking to disentangle the neural phase-locked component from that of the receptors ( CM ) . We demonstrated the validity of this method in cat in two respects: that it isolated neural components and that it yielded an upper-frequency limit of neural phase-locking close to that reported in single AN fibers [23] . In human and monkey , the same stimulus paradigm was used as previously developed in cat [48] . Some parameters were adjusted to optimize measurement time . Briefly , the neural signal is disambiguated from the CM by comparing the response to a tonal probe with the response to the same probe but preceded by a masker . To then extract only the neural phase-locked component and discard the CAP , the responses to 2 opposite stimulus polarities are subtracted from each other . All data were processed and analyzed with custom MATLAB ( The Mathworks ) scripts . To improve the response’s SNR , the uncorrelated background noise was reduced by averaging the response of many repetitions ( n > 127 ) , and multiple Q-values were obtained at every measured frequency . Nevertheless , due to time constraints ( 2–4 hours ) in awake humans and anesthetized monkey ( 4–6 hours ) , only a limited number of Q-values could be extracted in each subject . Therefore , the population data in human and monkey are not evenly distributed across frequencies . To cope with this unevenly distributed data and to minimize the influence of outliers , we obtained Robust-LOESS trend-values instead of mean-values . The LOESS is a nonparametric local regression function using weighted linear squares and a second-degree polynomial model . The Robust version , the RLOESS assigns lower weight to outliers in the regression . The weights are given by the bisquare function with 0 weight for deviations greater than 6 mean absolute deviations . In Fig1 , the RLOESS trend-values were obtained using the MATLAB ( The Matworks ) SMOOTH function from the averaged Q-values within a subject . Moreover , the trendlines were obtained using a RLOESS function with a span of 0 . 85 and by interpolating the results with smoothing splines ( FIT , MATLAB , option: “SmoothingSpline , ” parameter: 0 . 999 approximate cubic spline ) . The 10th and 90th percentiles of the RLOESS trendline were estimated using bootstrapping [49] ( n = 200 ) , which is a random resampling method with replacement . The RLOESS function and bootstrapping were performed on the average of repetitions ( same condition and experiment ) . For the neurophonic ( e . g . , Fig 3a and S8 Fig ) a similar approach was used to obtain the trendline . In this case , the LOESS function was used with span 0 . 55 , and the resulting trend values were connected by straight lines . The bootstrap standard error for the frequency limit of phase-locking was approximately 270 Hz for monkey and approximately 450 Hz for human . Because the absolute amplitude of the neurophonic is not only dependent on neural factors ( e . g . , cochlea-dependent spatiotemporal summation , electrode contact impedance , etc . ) , it can differ between subjects . Therefore , the individual data were first normalized to their maximum value before the application of bootstrapping . | The coding of sounds by the cochlea depends on two primary properties: frequency selectivity , which refers to the ability to separate sounds into their different frequency components , and phase-locking , which refers to the neural coding of the temporal waveform of these components . These properties have been well characterized in animals using neurophysiological recordings from single neurons of the auditory nerve ( AN ) , but this approach is not feasible in humans . As a result , there is considerable controversy as to how these two properties may differ between humans and the small animals typically used in neurophysiological studies . It has been proposed that humans excel both in frequency selectivity and in the range of frequencies over which they have phase-locking . We developed a technique to quantify these properties using mass potentials from the AN , recorded via the middle ear in human volunteers with normal hearing . We find that humans have unusually sharp frequency tuning but that the upper frequency limit of phase-locking is at best similar to—and more likely lower than—that of the nonhuman animals conventionally used in experiments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"chinchillas",
"medicine",
"and",
"health",
"sciences",
"ears",
"vertebrates",
"neuroscience",
"animals",
"mammals",
"primates",
"animal",
"models",
"inner",
"ear",
"computational",
"neuroscience",
"experimental",
"organism",
"systems",
"nerve",
"fibers",
"coding",
"mechanisms",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"monkeys",
"animal",
"cells",
"animal",
"studies",
"head",
"short",
"reports",
"macaque",
"cellular",
"neuroscience",
"rodents",
"eukaryota",
"anatomy",
"cell",
"biology",
"cochlea",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"computational",
"biology",
"amniotes",
"organisms"
] | 2018 | High-resolution frequency tuning but not temporal coding in the human cochlea |
The complex host-pathogen interplay involves the recognition of the pathogen by the host's innate immune system and countermeasures taken by the pathogen . Detection of invading bacteria by the host leads to rapid activation of the transcription factor NF-κB , followed by inflammation and eradication of the intruders . In response , some pathogens , including enteropathogenic Escherichia coli ( EPEC ) , acquired means of blocking NF-κB activation . We show that inhibition of NF-κB activation by EPEC involves the injection of NleE into the host cell . Importantly , we show that NleE inhibits NF-κB activation by preventing activation of IKKβ and consequently the degradation of the NF-κB inhibitor , IκB . This NleE activity is enhanced by , but is not dependent on , a second injected effector , NleB . In conclusion , this study describes two effectors , NleB and NleE , with no similarity to other known proteins , used by pathogens to manipulate NF-κB signaling pathways .
Enteropathogenic Escherichia coli ( EPEC ) belong to a group of pathogens defined by their ability to form “attaching and effacing” ( AE ) histopathology on intestinal epithelia . These pathogens employ their type III protein secretion system ( TTSS ) to inject ( translocate ) toxic proteins ( effectors ) into the host cell . The injected effectors subvert normal host cell functions to benefit the bacteria ( summarized in [1] ) . To date , 21 effectors or putative effector genes have been described for EPEC . Six of them are encoded in the LEE region that also encodes the TTSS structural genes , whereas the other 15 effector genes are distributed within three prophages and three insertion elements ( IE ) [2] . Upon infection , bacterial PAMPs ( pathogen-associated molecular patterns ) including LPS , flagellin , lipoproteins , and CpG DNA stimulate host cell Toll-like receptors ( TLRs ) in the host cells , leading to a formidable immune response via the activation of the transcription factor NF-κB [3] , [4] . NF-κB comprises a family of closely related transcription factors that play a key role in the expression of genes involved in inflammation , immune , and stress responses . NF-κB is a collective term used for homo- and heterodimeric complexes formed by the Rel/NF-κB proteins . In mammals , five of such proteins are known: RelA ( p65 ) , RelB , c-Rel , p50 ( NF-κB1 ) , and p52 ( NF-κB2 ) . Under nonstimulating conditions , NF-κB is retained in the cytoplasm through its association with inhibitory proteins ( IκBs ) . A variety of signaling pathways activate IκB kinases ( IKK ) to phosphorylate IκB , leading to its ubiquitination and degradation by the proteasome . This allows translocation of NF-κB to the nucleus , activation of NF-κB-regulated genes , and establishment of an inflammatory response [5] , [6] . Previous reports have suggested that during infection , EPEC manipulate NF-κB-mediated inflammation . Initially , it was shown that EPEC activate NF-κB by a TTSS-dependent mechanism [7] , [8] , but later , it was reported that the TTSS is not required and that EPEC activate NF-κB via a TTSS-independent mechanism , presumably by activation of TLRs [9] , [10] , [11] , [12] . Moreover , some reports showed that EPEC actually repress NF-κB activation by a TTSS-dependent mechanism [13] , [14] . Taken together , these reports suggest that EPEC first elicit NF-κB activation by a TTSS-independent mechanism and subsequently , it utilize the TTSS mechanism to mediate TTSS-dependent NF-κB-repression . However , the major gap in the above hypothesis is that the putative effector that presumably represses NF-κB activation has never been identified . In this report we confirm that EPEC block NF-κB activation via a TTSS-dependent mechanism and show that the NleE effector is necessary and sufficient to block NF-κB activation via inhibition of IκB phosphorylation and thus induces its stabilization . In addition , we show that a second effector , NleB , is required for better repression of NF-κB activation , suggesting that the function of NleB is related to that of NleE .
The ability of EPEC to either inhibit or induce NF-κB activation is controversial . Therefore , we re-examined this point using HeLa cells as host cells and IκB stability as a read-out for NF-κB activation . Importantly , TNFα treatment strongly stimulates IκB degradation in these cells , but under the used experimental conditions they exhibited minimal IκB degradation upon exposure to the PAMPs of the infecting EPEC . This allowed the uncoupling of infection and NF-κB activation . HeLa cells were infected with EPEC culture for 3 h , during which the bacteria injected the TTSS effectors into host cells . The infected cells were then treated with 10 ng/ml TNFα to activate NF-κB and at different time points post TNFα-induction , cellular lysates were subjected to western analysis . The results show that TNFα treatment induced rapid degradation of IκB in uninfected cells or cells infected with EPEC TTSS-deficient mutant ( escN::kan ) . In contrast , IκB in cells infected with wild-type EPEC remained stable ( Fig . 1A ) . We next tested whether the stabilization of IκB by EPEC was associated with inhibition of NF-κB translocation to the nucleus . To monitor NF-κB activation , we used a reporter cell line ( AGS SIB02 ) stably expressing the NF-κB subunit p65 fused to GFP . Cells were infected with wild-type EPEC or , as a negative control , with EPEC TTSS mutant ( escV::kan ) . After 3 h of infection , cells were washed , induced with TNFα , and at 15 , 45 , 60 , and 75 min post TNFα-induction , the cells were fixed , stained with Hoechst 33342 , and analyzed by automated microscopy . Importantly , whereas wild-type EPEC repressed p65-GFP translocation to the nucleus , the TTSS escV mutant was strongly attenuated in this activity ( Fig . 1B ) . To validate the above microscopic analysis we carried out identical experiment , but instead of using microscopy to analyze the cells we fractionated the cells into cytoplasmic and nucleus fractions and determined the amount of p65 in the different fractions by immunoblot using anti-p65 antibody . The results were in agreement with the microscopic analysis: wild type EPEC , but not the escV mutant , blocked translocation of p65 to the nucleus ( Fig . 1C ) . Thus supporting the notion that EPEC inhibit NF-κB activation by a TTSS-dependent mechanism [13] , [14] , and suggesting that EPEC deliver into infected cells one or more effectors that inhibits NF-κB activation . NleH has been proposed as such an effector since it is similar to OspG , a Shigella effector that inhibits NF-κB activation [15] . However , we found that an EPEC strain , in which both nleH alleles were deleted , still inhibited IκB degradation , similarly to wild-type EPEC ( Fig . 1D ) , suggesting that NleH is not required for blocking IκB degradation under the experimental conditions used by us . To identify putative effector ( s ) that block IκB degradation , we bioinformaticly compared the genome of EPEC to that of non-pathogenic E . coli K12 and identified large EPEC-specific regions that contain , or possibly contain , effector genes . Based on this comparison , we constructed a set of 15 EPEC strains , each deleted of one EPEC-specific large chromosomal region ( Table 1 ) . Altogether , 770 EPEC-specific ORFs were deleted . We then tested the capacity of each of the deleted strains to inhibit IκB degradation upon TNFα treatment . One of the strains , deleted of the IE6 region [2] , could not inhibit IκB degradation ( data not shown ) . Further systematic deletion analysis defined two effector-encoding genes , nleB and nleE , required for stabilizing IκB ( Fig . 2A , 2B and data not shown ) . Deletion of nleE strongly reduced the bacteria's capacity to stabilize IκB , but a complete deficiency in IκB stabilization was observed only in the strain deleted of both nleB and nleE ( Fig . 2B ) . To corroborate the notion that NleE is required for IκB stabilization , we complemented a strain deleted of the nleBE region with plasmids containing nleB , nleE , or nleBE . We found that expression of NleE , but not of NleB , partially restored EPEC's capacity to stabilize IκB ( Fig . 2C ) . Importantly , full IκB protection was achieved in strains expressing both NleB and NleE ( Fig . 2C ) . A mutant expressing only NleB showed only low level of IκB protection ( Fig . 2C ) . Taken together , these results suggest that NleB and NleE , located at the IE6 region , are necessary for stabilizing IκB and that this activity is contributed mainly by NleE ( Fig . 2B ) . We therefore focused our attention on NleE . EPEC encode two very similar nleE alleles . One allele , identified in our screen , is located in the IE6 region and the other is in the IE2 region [2] . We initially found that deletion of the IE6 region , but not of the IE2 region , caused deficiency in inhibition of IκB degradation ( data not shown ) . However , the two proteins , NleEIE2 and NleEIE6 , are identical , apart from an internal deletion of 56 residues in NleEIE2 ( Fig . S1 ) , and this similarity between the two proteins urged us to determine the activity of each of the two proteins . We first tested their ability to complement IκB destabilization in a strain deleted of nleEIE6 . To this end , we expressed each of them on a plasmid carrying an identical promoter and ribosomal binding site . Results showed that only NleEIE6 , but not NleEIE2 , was able to attenuate IκB degradation ( Fig . 3A ) . These results indicate that NleEIE2 is either not active in the host cell or is not translocated into the host cell . To differentiate between these two possibilities , we used the above mentioned plasmid where both proteins were fused to the β-lactamase translocation reporter protein , BlaM . The plasmids were introduced into EPEC and the ability to translocate them into infected cells was tested . We found that both NleEIE2-BlaM and NleEIE6-BlaM were expressed at similar levels in the bacteria ( Fig . 3B ) . Importantly , however , only NleEIE6 was translocated into the host cell ( Fig . 3C ) , suggesting that NleEIE2 is a cryptic effector . Cumulatively , these results define NleEIE6 , but not NleEIE2 , as the effector needed for inhibition of IκB degradation . Therefore , in this report the term “NleE” specifically refers to “NleEIE6” . To corroborate the notion that NleE is required for inhibition of NF-κB activation , we examined whether the nleE mutant is deficient in blocking the translocation of the p65 NF-κB subunit to the nucleus upon TNFα treatment . Because of the high similarity between the IE2 region and the IE6 region , we made the specific mutants ( ΔnleEIE6 , ΔnleEBIE6 ) in a strain deleted of its IE2 region ( ΔIE2 ) . Therefore , these specific strains were compared to their parental strain ( indicated as WTΔIE2 in the figures ) . It should be emphasized that the ΔIE2 mutant exhibited the wild-type phenotype in all the assays used in this study . AGS SIB02 cells expressing p65-GFP were infected with the parental strain WTΔIE2 or with the corresponding mutants: ΔescV , ΔnleE , ΔnleBE , and ΔnleE complemented with a plasmid expressing nleE . After 3 h , cells were TNFα-induced for 30 min , stained with Hoechst 33342 and analyzed by automated microscopy . The results show that whereas the wild type repressed ∼90% of the p65 translocation to the nucleus , the TTSS escV mutant was attenuated , exhibiting only ∼50% repression ( Fig . 4A ) . These results indicate that EPEC inhibit p65 translocation by both TTSS-independent and TTSS-dependent pathways . Importantly , the nleE and nleBE mutants were as deficient as the escV mutant in blocking p65 translocation . Moreover , complementation of the nleE mutant with the wild-type nleE allele restored the bacteria's full capacity to inhibit the p65 translocation ( Fig . 4A ) . These results support the hypothesis that the TTSS-mediated inhibition of p65 translocation is NleE-dependent . In addition , our results suggest that a putative TTSS-independent mechanism might function in parallel to NleE to inhibit p65 translocation . To further substantiate our results , we used IL-8 expression as an additional read-out for NF-κB activation . Briefly , HeLa cells were infected with different EPEC strains or remained uninfected . Then , cells were washed and treated for 3 h with TNFα and gentamycin , to kill the remaining bacteria . RNA was then extracted from the cells and the amount of produced IL-8 mRNA was measured by real time PCR . In comparison to non infected cells or cells infected with EPEC escV mutant , both wild type and the ΔIE2 mutant exhibit a ∼100 fold repression of IL8 expression ( Fig . 4B ) . In contrast , the nleE mutant exhibited a partial , less then 10 fold , repression of IL8 expression and this was moderately complemented by plasmid expressing native NleE ( Fig . 4B ) . A more severe deficiency in repression of IL8 expression was exhibited by a double mutant nleBE ( Fig . S3 ) . Furthermore , a plasmid expressing nleBE restored IL8 repression to that seen in wild type EPEC ( Fig . S3 ) . Upon testing the amount of secreted IL8 protein instead of production of IL8 mRNA , similar results were obtained ( Fig . S2 ) . Taken together these results show that ( i ) NleE is required for full inhibition of IL-8 expression , ( ii ) NleB also contributes to this repression and iii ) a putative TTSS effector ( s ) , other then NleB and NleE might function in parallel to inhibit IL-8 expression . The IL8 expression assay was found to be much more sensitive then testing translocation to the nucleus or the IkB degradation assay . This is probably since the latter are very transient events while the mRNA tends to accumulate , increasing the signal/noise ratio . Interestingly , using the IL8 expression assay we found that infection with the escV mutant was sufficient to induce IL8 expression in HeLa cells , albeit not as strong as that induced by TNFα ( data not shown ) . This activation is possibly via the activity of flagellin , LPS or other PAMPs . We thus next asked whether NleE also inhibits the EPEC-induced IL8 expression . To this end we repeated the experiment described in Fig . 4B , but TNFα was omitted . We found that even the non infected cells produce certain levels of IL8 mRNA , but upon infection with EPEC escV mutant we observed a ∼10 fold increase in IL8 expression ( Fig . 4C ) . In contrast , the EPEC wild type ( or the ΔIE2 mutant ) exhibited strong repression of the EPEC-induced IL8 expression . Importantly , the nleE mutant exhibited only a partial capacity to repress the self-induced IL8 expression . Similar results where observed when we used the double mutant nleBE instead of nleE mutant ( Fig . S3 ) . However , both the nleE or the nleBE , mutants were not as deficient in IL8 repression as the escV mutant ( Fig . 4C and S3 ) . Thus , we predict that additional putative effector might function in parallel to NleB and NleE to repress IL8 expression . In conclusion , our results clearly show that i ) EPEC mediate a TTSS-dependent repression of self-induced IL8 expression; and ii ) NleE is required for full repression of the EPEC-induced IL8 expression . We next examined whether NleE is sufficient for inhibition of NF-κB activation in the absence of the infecting bacteria and other putative effectors . To this end , we constructed a vector expressing mCherry fused to NleE ( mCherry-NleE ) and used it for transient transfection of HeLa cells . Untransfected cells ( Fig . 5A ) or cells transfected with either the mCherry-NleE vector or a vector expressing mCherry alone ( Fig . 5B ) were stimulated with TNFα for 1 h , or remained untreated . Next , these cells were fixed , stained with anti-p65 antibody , and analyzed by fluorescent microscopy to determine both the ability of the transiently expressed NleE to inhibit TNFα-induced migration of p65 to the nucleus and to determine its localization in the expressing cells . The TNF treatment induced strong migration of p65 to the nucleus in the untransfected cells ( Fig . 5A ) , and in cells transfected with the negative control vector ( Fig . 5B two upper panels and 5C ) . Importantly , the transiently expressed mCherry-NleE induced a strong inhibition of p65 translocation to the nucleus ( Fig . 5B two lower panels and 5C ) . The expressed mCherry and mCherry-NleE were similarly distributed in the cells , predominantly in the cytoplasm ( Fig . 5B ) . These results indicate that NleEIE6 is sufficient for inhibition of NF-κB migration to the nucleus presumably by IκB stabilization . Similar analysis using NleEIE2 instead of NleEIE6 , show that NleEIE2 lost the ability to block p65 translocation to the nucleus ( Fig . S4 ) , highlighting the importance for NleE activity of the region between residues 49–115 , which is deleted in NleEIE2 ( Fig . S1 ) . Different NF-κB activating pathways converge at the level of IKK phosphorylation , which subsequently leads to IκB phosphorylation , targeting it to ubiquitination and proteasome-mediated degradation [6] . We thus tested whether NleE inhibits the TNFα-induced IκB phosphorylation . Cells were infected with different EPEC strains followed by TNFα treatment . The levels of IκB and phospho-IκB were then determined by immunoblot analysis with the appropriate antibodies and the relative accumulation of unphosphorylated IκB was determined . For a negative control , we used cells infected with the escN mutant , which cannot stabilize IκB ( Fig . 1A ) . Indeed , in cells infected with this mutant we noted increased IκB phosphorylation followed by its degradation . However , the addition of proteasome inhibitor ( MG132 ) resulted in accumulation of phosphorylated IκB ( Fig . 6A ) . As a parental strain , we used the ΔIE2 strain ( WTΔIE2 ) , which , like wild-type EPEC , efficiently protected IκB from degradation ( Fig . 1A and 6A ) . Importantly , the accumulated IκB in these cells was mostly unphosphorylated . In contrast , the corresponding nleE mutant failed to induce accumulation of unphosphorylated IκB , exhibiting a phenotype similar to that of the escN mutant ( Fig . 6A ) . Taken together , these results indicate that wild-type EPEC stabilizes IκB by preventing its phosphorylation and that NleE is required for this activity . Indeed , complementing the ΔnleE mutant with a plasmid expressing NleE restored the bacteria's capacity to induce the accumulation of unphosphorylated IκB ( Fig . 6A ) . These results suggest that one can restore the inability of the ΔnleE mutant to prevent IκB degradation by two alternative approaches: ( i ) by treatment with proteosome inhibition , to inhibit phospho-IκB degradation , or ( ii ) by complementation with a plasmid expressing nleE , to block IκB phosphorylation . To compare the efficiency of these two treatments , we infected HeLa cells with the ΔIE2 strain ( WTΔIE2 ) , ΔnleE mutant , or with the ΔnleE mutant complemented either by proteosome inhibitor ( MG132 ) treatment upon TNFα induction , or by a plasmid expressing nleE . The results show that both treatments similarly stabilized the IκB . However , the first treatment led to a strong phosphorylation of the accumulated IκB whereas when NleE was added , the accumulated IκB remained unphosphorylated ( Fig . 6B ) . These results further support the notion that NleE stabilizes IκB by inhibiting its phosphorylation . The signaling pathways induced by the TNF receptor ( TNFR ) is different from that induced by the IL1 or TLR receptors , but both converge at the level of IKK activation by TAK1 ( Fig . 6C , [16] ) . The inhibition of the self-induced IL8 expression by NleE ( Fig . 4C ) , is hinting that NleE functions downstream to the pathways converging point . To directly test this prediction we tested whether EPEC is capable of inhibiting IL1β-induced degradation of IκB . Importantly , we found that wild type EPEC , but not the nleE mutant , inhibited the IL1β-induced IκB degradation ( Fig . 6D ) . These results confirmed that NleE functions downstream to the signaling converging point . We next tested whether NleE can block the phosphorylation and thus activation of IKKβ . To this end we extracted proteins from cells , which were infected with different strains and then treated with TNFα or IL1β as indicated ( Fig . 6E ) . The extracted proteins were subjected to western analysis using anti-IKKβ , anti-phospho-IKK , anti-IκB and anti-phospho-IκB antibodies . The results show that , treatment with either TNFα or IL1β induced IκB and IKK phosphorylation in non infected cells or cells infected with the escV mutant . We also found that wild type EPEC , but not the nleE mutant , inhibited this IKK phosphorylation . The same inhibition is noted for the IkB phosphorylation , in the wildtype strain , However , due to IkB degradation , less protein is noted and thus its phosphorylation cannot be seen ( Fig . 6E ) . Complementation with plasmid expressing wild type nleE allele only partially , but consistently restored the inhibition of IKKβ phosphorylation ( Fig . 6E ) . These results suggest that NleE block activation of IKKβ . Taken together our results indicate that NleE blocks the NF-κB signaling cascade downstream to the converging point of the TNFα and IL1β signaling pathways , but upstream to IκB phosphorylation , possibly by direct blocking TAK1 or IKKβ activation .
In this report we showed that NleE of EPEC stabilizes the NF-κB inhibitor , IκB , via inhibition of its phosphorylation , thereby preventing NF-κB signaling . This activity of NleE was discovered using an unbiased screen of EPEC strains deleted of most of the EPEC-specific genes . We showed that an nleE mutant is deficient in blocking IκB phosphorylation and in preventing its degradation . Moreover , an nleE mutant was attenuated in blocking TNFα-induced NF-κB migration to the nucleus as well as in IL-8 expression and secretion . These abilities were restored to the mutant upon complementation with a plasmid expressing the wild-type nleE allele . Importantly , we showed that NleE expressed in HeLa cells blocks NF-κB translocation to the nucleus upon TNFα treatment . Taken together , our findings indicate that NleE is sufficient to inhibit NF-κB signaling by blocking IκB phosphorylation . Further analysis suggest that NleE blocks the NF-κB signaling cascade downstream to the converging point of the TNFα and IL1β signaling pathways , but upstream to IκB phosphorylation , possibly by directly blocking TAK1 or IKKβ activation . We also show that NleB enhances NleE activity . The nleE mutant still showed residual inhibition of IκB degradation , which was eliminated upon further deletion of nleB . Moreover , complementation of the nleBE double mutant with a plasmid expressing nleBE was more efficient than a plasmid expressing only nleE . These results suggest that NleB plays a role in IκB stabilization . Similar results were obtained when expression of IL8 was used as a readout for inhibition of the NF-κB signaling . The mechanism underlying NleB function is not yet apparent . Nevertheless , the notion that NleB and NleE function together is supported by the facts that nleE form a putative bicistronic operon with nleB and that nleE is consistently associated with nleB in natural isolates of diarrheagenic EPEC [17] . Other isolates of EPEC , EHEC , and C . rodentium carry nleB and nleE alleles almost identical to nleBEIE6 investigated in this study ( Supplemental material Fig . S1 ) . We predict that all NleBE proteins function similarly . The phenomenon of effectors functioning in parallel is common in EPEC [1] . Interestingly , the TTSS mutants ( escV ) and ΔnleBE were only partially deficient in blocking the TNFα-induced migration of NF-κB to the nucleus , suggesting that an additional TTSS-independent mechanism might function in parallel to NleBE to inhibit translocation of NF-κB to the nucleus ( Fig . 7 ) . This activity might be related to damping of the signal due to continuous exposure to PAMPs like flagellin or LPS . Interestingly , the TTSS escV mutant was completely deficient in inhibition of IL8 expression , while the nleBE mutant was only partially deficient in this function . We therefore predict that EPEC encode additional putative effector ( s ) that function in parallel to NleBE by blocking IL-8 expression ( Fig . 7 ) . Where as in this study we showed that NleE represses NF-κB activation , a recent report suggested exactly the opposite , i . e . NleE and OspZ , an NleE-homolog encoded by Shigella , activates NF-κB signaling [18] . The discrepancy between the two studies might result from the use of different cell lines . In their study , Zurawski et al . [18] used cell lines that activate NF-κB upon sensing of either TNFα or bacterial PAMPs including LPS . The latter might complicate the interpretation of the results . To avoid such complications , we used in this study HeLa and AGS cells , which are inert to LPS . Moreover , we adjusted the experimental conditions used in this study , so that the NF-κB activation by other bacterial PAMPs was insignificant . This facilitated the uncoupling of effector injection and NF-κB activation , which occurred only upon addition of TNFα . An alternative , although less likely explanation for the inconsistency between the two studies is that NleE functions differently in different types of cell or different cell lines . Like EPEC , Yersinia also employs an injected effector , YopJ , to block IκB phosphorylation . YopJ is an acetyl transferase that acetylates critical IKKβ residues and thus prevents its activation [19] . Although both NleE and YopJ block IKKβ phosphorylation , they are very different in sequence , which probably reflects functional differences . We are currently investigating the NleE's mode of action . Other effectors that interfere with NF-κB function include the Salmonella SspH , and Shigella IpaH9 . 8 , which are targeted to the host cell nucleus and inhibit NF-κB-dependent transcription [20] . Shigella also uses OspG , which inhibits ubiquitination of phospho-IκB [15] . Interestingly , EPEC encode an OspG homolog , NleH . However , we found that an EPEC strain , mutated in its two nleH alleles , still inhibit IκB degradation . Yet , our analysis indicates that additional TTSS effector ( s ) inhibits NF-κB signaling ( Fig . 7 ) and we can not exclude the possibility that this putative effector is nleH . The role of NleE in EPEC virulence was not tested , since an animal model is not yet available , but it was tested with C . rodentium . Importantly , NleE was found to be required for full virulence of C . rodentium upon infection of wild-type mice , but this requirement was diminished upon infection of mice deficient in TLR4 [21] , [22] . Our results revealed the rationale behind this intriguing phenomenon . We argue that NleE-mediated NF-κB repression is no longer needed if the host itself is deficient in TLR4/LPS-induced NF-κB signaling . In conclusion , we show that NleE blocks IκB phosphorylation by IKK and thus it inhibits NF-κB signaling . We also show that NleB enhances NleE's activity and that EPEC probably use additional mechanisms to interfere with other constituents of the NF-κB signaling pathway . This is presumed to have multiple consequences on the course of EPEC infection and the maturation of both innate and adaptive host immune response .
Bacterial strains , plasmids , and primers used in this study are listed in Table S1 , S2 , and S3 , respectively ( Supplemental material ) . Deletions in the EPEC chromosome were constructed using the primers listed in Table S3 , as described [23] . For bacterial expression , the genes were cloned in the pSA10 expression vector as described [24] . In most cases the leakiness of the Tac promotor was enough for gene expression . When indicated , IPTG ( 0 . 01mM ) was added . For the cloning procedure , genes were amplified by RCR using the primers listed in Table S3 . Formation of plasmid-borne nleE-blaM fusions were carried out as described [25] using the plasmid pCX341 and primers as listed in Table S3 . The nleEIE2 DNA was amplified from EPEC ΔIE6 mutant and that of nleEIE6 from EPEC ΔIE2 mutant . For formation of mCherry fusions , the EGFP gene of pEGFP-N1 ( Clonetech ) was excised from the plasmid using the NotI and BamHI and replaced by mCherry taken from pREST-mCherry , resulting in pMS2841 . This plasmid was transformed into pSC4141 by introduction of a unique scaI site at the mCherry 3′ , eliminating its stop codon in the process . This was done using QuikChange site-directed mutagenesis kit ( Stratagene #200518-5 ) and primers listed at Table S3 . This plasmid was further used to create the transcriptional fusion: mCherry-nleEIE6-6his ( pSC4144 ) and mCherry-nleEIE2-6his ( pSC4350 ) , using primers listed at Table S3 . HeLa cells ( 9×105 ) in 4 cm plates were infected with a 1∶100 dilution in DMEM of bacteria grown overnight statically at 37°C ( multiplicity of infection , MOI∼1∶100 ) . Following 3 h infection in 5% CO2 , at 37°C , the medium was replaced with fresh DMEM with or without either 10 ng/ml TNFα for 40 min ( or 20 ng/ml IL-1β for 20 mins ) . When indicated , the infecting bacteria were supplemented with 0 . 01 mM IPTG at 1 . 5 h post inoculation . To terminate the infection , cells were washed with 3 ml of cold TBS ( 20 mM Tris-HCl , pH 7 . 4 , 150 mM NaCl ) , scraped with 1 ml of cold TBS , collected and centrifuged , ( 800g , 2 min at 4°C ) . The pellet was resuspended in 40 µl lysis buffer ( 0 . 5% Triton-×100 , 20mM Tris-HCl pH 7 . 2 , 0 . 2 mM VO4 , 10 mM NaF , 30 µl of complete inhibitor Roche ) and centrifuged ( 20 , 000g , 3 min at 4°C ) . Supernatant was either transferred for protein quantification assay ( BCA assay ) or to a tube with loading dye ( LDS sample buffer , NuPAGE ) , boiled for 10 min , and then centrifuged ( 20 , 000g , 3 min ) . Samples were quantified using bicinchoninic acid ( BCA ) and copper sulfate . Equal protein concentration for each sample was then loaded on SDS-PAGE gel , transferred to PVDF membrane , and reacted with antibodies against IκB ( 1∶1000 ) , Tubulin , as a loading verification control ( 1∶2500 ) , or phospho-IκB ( 1∶1000 , Cell Signaling ) . When indicated , 20 mM MG132 ( 1∶1000 ) was used . Protein band density was quantified using Tina software ( version 2 . 09 ) and the percentage of the unphosphorylated IκB was determined by calculating the relative phosphorylated IκB out of the total IκB shown for each lane . IKKβ analysis was done as described for IκB except that induction time with TNFα was reduced to 10 min and with IL1β it remained 20 min . IKK detection was preformed by western blot analysis using anti-IKKβ antibody ( 1∶1000 , Cell Signaling Technology , #2684 ) and Phospho-IKKα ( Ser180 ) /IKKβ ( Ser181 ) antibody ( 1∶1000 , Cell Signaling Technology , #2681S ) Generation of the p65-GFP-expressing cell line and the automated image analysis to quantify translocation of p65-GFP is described elsewhere ( Bartfeld et al . , submitted ) . Briefly , SIB02 cells are AGS cells lentivirally transduced to express p65-GFP . SIB02 cells , seeded in 96-well-plates , were inoculated with EPEC strains at MOI 1∶100 , incubated for 3 h and subsequently activated by 10 ng/ml TNFα . After an appropriate incubation time , cells were fixed using 100% ice-cold methanol and stained with Hoechst 33342 ( 2 µg/ml ) . Images of ∼200 cells were acquired using automated microscopy ( Scan∧R , Olympus ) and translocation of p65-GFP to the nucleus was subsequently quantified using Scan∧R image analysis software ( Olympus ) as described ( Bartfeld et . al . , submitted ) . Cells with nuclear p65-GFP above the defined threshold were termed “active” and the percentage of active cells per well was calculated . HeLa cells ( 2 . 8×106 ) were seeded in 10 cm plates . The next day , the cells were infected with EPEC for 3 . 5 h as described . Cells were then washed , treated with 20 ng/ml TNFα in DMEM for 30 min . , washed with cold PBS , scrapped , transferred to Eppendorf tubes and centrifuged ( 5 mins , 660 g , 4°C ) . Then , the pellet was resuspended in 7 times the volume of Hypotonic Lysis Buffer ( HLB , 10mM HEPES pH 7 . 6 , 0 . 1mM EDTA , 0 . 1 mM EGTA , 2mM DTT , 10mM KCl , 1mM PMSF , 0 . 75mM Sperimidine , 0 . 15mM Sperimide , 20mM PNPP , 1µM okadaic acid and 5µg/ml protease inhibitor ) , incubated on ice for 15 mins and then 0 . 2% NP40 was added gently following gentle mixing for several minutes . The lysate was then centrifuged ( 5 mins , 2600 g , 4°C ) , the supernatant ( cytoplasmic fraction ) was recovered and the pellet ( nuclear fraction ) was washed with HLB once and then resuspended in 100 µl Nuclear Extraction Buffer ( NEB , 210 mM HEPES pH 7 . 6 , 0 . 2 mM EDTA , 2 mM EGTA , 0 . 5 mM DTT , 25% Glycerol , 0 . 42 M NaCl , 20 mM glycerophosphate , 29 mM PNPP , 1 µM okadiac acid , 1 mM NaVO4 , 5 µg/ml protease inhibitor , 0 . 75 mM Sperimidine , 0 . 15 mM Sperimide ) . The nuclear lysets were then vortexed , mixed vigorously ( 1400 rpm , 30 min . , 4°C ) and clarified ( 20 , 000 g , 10 min , 4°C ) . Protein concentrations were determined ( BCA kit , Sigma ) , adjusted and the extracts were used for western analysis using anti-NF-κB p65 antibodies ( Santa Cruz , SC372 ) . The quality of the fractionation was confirmed using tubulin as a cytoplasmic marker and fibrillarin as a nuclear marker . To determine translocation levels , overnight cultures of wild-type EPEC containing plasmids expressing NleE-BlaM were diluted 1∶50 in DMEM and used to infect HeLa cells for 3 h . Cells were then washed and stained with CCF2 for 2 . 5 h as described [26] , washed in cDMEM , excited at 405 nm , and then emission at 465 nm and 535 nm was recorded ( SPECTRAFluor , TECAN ) . The amount of translocation was determined as described [26] . As a negative control , we used EPEC expressing unfused BlaM ( Vector ) . To determine expression levels , the unattached bacteria were harvested , washed , and lysed by repeated freezing and thawing in PBS containing 1 mM EDTA , 1 mg/ml lysozyme , and 0 . 1% Triton-×100 . The BlaM activity in the lysate was determined using nitrocefin as substrate and the rate of product accumulation per number of bacteria ( OD 600 ) was determined as described [25] . HeLa cells ( 7×105 ) in 6 wells plates were inoculated with a 1∶100 dilution in DMEM of bacteria grown overnight statically at 37°C ( multiplicity of infection , MOI∼1∶100 ) and incubated for 3 h ( 5% CO2 , 37°C ) . To terminate the infection and induce IL8 expression , the medium was replaced with fresh DMEM supplemented with 2% FCS , 100ug/ul gentamicin and with or without 10 ng/ml TNFα and incubated for additional 3 h . Cells were than washed with 2 ml of cold TBS ( 20 mM Tris-HCl , pH 7 . 4 , 150 mM NaCl ) , scraped with 1 ml of cold TBS , collected and centrifuged , ( 800 g , 2 min , 4°C ) . RNA was extracted using the MasterPure Complete DNA and RNA Purification Kit ( EPICENTRE Biotechnologies ) and used to synthesize cDNA with the Verso cDNA kit ( Thermo scientific ) . hHPRT transcript levels were used to normalize total RNA levels in samples . Real time analysis was than conducted using Absolute Blue QPCR SYBR Green ( Thermo scientific ) in a real-time cycler ( Rotor-Gene 6000 , Corbett ) . HeLa cells were transfected using ExGen500 ( Fermentas ) , as recommended by the manufacturer , with 1 µg of pMS2841 ( pmCherry ) , pSC4144 or pSC4350 ( pmCherry-nleEs ) or were not transfected . After 24 h , the medium was replaced with fresh DMEM containing , or not containing , 10 ng/ml TNFα . After 1 h , cells were fixed ( 3 . 7% PFA in PBS for 10 min and washed with PBS ) , perforated ( with 0 . 25% Triton-X100 in PBS for 10 min and washed twice with PBS ) and blocked ( 2% BSA in TBS ) at 4°C for 16 h . Cells were then stained using anti-p65 ( SC109 , Cell Signaling ) antibodies ( 1∶300 in TBS ) overnight and further stained with CY-488 goat anti-rabbit ( Cell Signaling ) ( 1∶1000 in TBS ) for 1 h . Slides were analyzed by fluorescent microscopy . | The innate immune system senses intruding pathogens and in response , mounts an inflammatory reaction . Essential for this response is the activation of the transcription factor NF-κB , which mediates reprogramming of gene expression in the host . The bacteria Escherichia coli is usually a non-pathogenic resident of our intestinal track . Some E . coli strains , however , cause disease or food poisoning; one of these pathogenic strains is enteropathogenic E . coli ( EPEC ) . This pathogen employs a syringe-like organelle , termed type three secretion system ( TTSS ) , to inject into the intestinal host cell a battery of toxic proteins termed effectors . We found that two of the effectors that EPEC injects into the host cell upon infection block the activation of NF-κB and thus interfere with the host immune response . These findings elucidate the intricate cross-talk between the host immune system and the pathogen . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunology/immune",
"response",
"microbiology/innate",
"immunity"
] | 2010 | The Type III Secretion Effector NleE Inhibits NF-κB Activation |
Drosophila melanogaster Held Out Wings ( HOW ) is a conserved RNA–binding protein ( RBP ) belonging to the STAR family , whose closest mammalian ortholog Quaking ( QKI ) has been implicated in embryonic development and nervous system myelination . The HOW RBP modulates a variety of developmental processes by controlling mRNA levels and the splicing profile of multiple key regulatory genes; however , mechanisms regulating its activity in tissues have yet to be elucidated . Here , we link receptor tyrosine kinase ( RTK ) signaling to the regulation of QKI subfamily of STAR proteins , by showing that HOW undergoes phosphorylation by MAPK/ERK . Importantly , we show that this modification facilitates HOW dimerization and potentiates its ability to bind RNA and regulate its levels . Employing an antibody that specifically recognizes phosphorylated HOW , we show that HOW is phosphorylated in embryonic muscles and heart cardioblasts in vivo , thus documenting for the first time Serine/Threonine ( Ser/Thr ) phosphorylation of a STAR protein in the context of an intact organism . We also identify the sallimus/D-titin ( sls ) gene as a novel muscle target of HOW–mediated negative regulation and further show that this regulation is phosphorylation-dependent , underscoring the physiological relevance of this modification . Importantly , we demonstrate that HOW Thr phosphorylation is reduced following muscle-specific knock down of Drosophila MAPK rolled and that , correspondingly , Sls is elevated in these muscles , similarly to the HOW RNAi effect . Taken together , our results provide a coherent mechanism of differential HOW activation; MAPK/ERK-dependent phosphorylation of HOW promotes the formation of HOW dimers and thus enhances its activity in controlling mRNA levels of key muscle-specific genes . Hence , our findings bridge between MAPK/ERK signaling and RNA regulation in developing muscles .
Regulation of gene expression at the level of RNA is often mediated through the activities of RNA-binding proteins ( RBPs ) , which control different aspects of RNA metabolism of target genes [1] , [2] . Drosophila melanogaster Held Out Wings ( HOW ) is an RBP that belongs to a family of evolutionarily conserved “Signal Transduction and Activation of RNA” ( STAR ) proteins [3] . STAR family members control a wide range of tissue differentiation processes . For example , in mammals , Sam68 controls spermatogenesis [4] , and Quaking ( QKI ) regulates myelination by Schwann cells and oligodendrocytes [5] , [6] , [7] as well as muscle fiber maturation in Zebrafish [8] . In C . elegans , the STAR protein GLD-1 promotes germ cell differentiation [9] , while ASD-2 is required for alternative splicing [10] . HOW , the Drosophila protein orthologous to mammalian QKI , is highly expressed in muscles , tendons [11] , [12] , [13] , [14] and glial cells [15] , [16] , where it plays an essential role during development by controlling the mRNA levels of an array of target genes [17] . HOW performs various activities on its target RNAs: it facilitates the alternative splicing of stripe A , a transcription factor essential for tendon cell maturation [18] , and mediates specific splicing of the septate junction constituent , nrxIV , thereby controlling glial cell maturation [15] . It also functions by reducing mRNA levels of various targets . For example , during gastrulation , HOW-dependent downregulation of cdc25/string , a cell cycle promoting phosphatase , is essential to inhibit cell division in invaginating mesodermal cells [19] . Structurally , HOW contains a single maxi-KH RNA binding motif that is flanked by two additional conserved domains , QUA1 and QUA2 [14] . While the QUA2 motif , located C terminally to the KH domain , takes part in RNA-binding and contributes to the specificity of RNA recognition [20] , [21] , [22] , the QUA1 motif , located N terminally to the KH domain , was shown to mediate protein dimerization in GLD-1 , Sam68 and QKI [23] , [24] , [25] . Notably , despite the fact that this domain is not essential for RNA-binding , its deletion in GLD-1 nevertheless reduces the affinity of the protein to the RNA binding motif TGE ( Tra-2 and GLI response element ) , by about ten-fold [26] , suggesting that dimerization of STAR proteins might enhance their affinity to RNA . To date , it is not clear what regulates the degree of STAR protein dimer formation . Consistent with their expression in a wide range of tissues during development , the activity of STAR proteins is highly regulated at distinct post-translational levels , including phosphorylation by various kinases ( reviewed in [27] ) . These modifications likely impinge on the activity , subcellular distribution , or the formation of protein complexes of specific STAR proteins . Phosphorylation of STAR proteins could couple regulation of RNA metabolism with distinct signaling cascades operating in a spatial and temporal restricted manner . While Tyrosine phosphorylation of the C terminal regions of both Sam68 [28] and QKI [29] has been established , Serine/Threonine ( Ser/Thr ) phosphorylation has yet to be demonstrated with respect to the QKI-subfamily of proteins . A more evolutionarily distant STAR protein , Sam68 , was shown to be Ser/Thr phosphorylated , both by Cdc2 [30] and by ERK1/2 in various culture lines , promoting differential activities . Specifically , Sam68 phosphorylation by ERK1/2 in a lymphoma cell line enhances its ability to regulate alternative splicing [31] , while in mouse spermatocytes and in HEK293 cells it induces cytoplasmic accumulation correlated with its association with polyribosomes [4] , [32] . We therefore postulated that ERK-dependent phosphorylation of STAR proteins might affect their activity in a tissue-specific manner . In the present manuscript , we show that the Drosophila STAR protein HOW is phosphorylated on conserved Thr residues . Importantly , we demonstrate that in embryos in vivo , a particular HOW isoform is phosphorylated on Thr , in a tissue-specific manner . Moreover , we identify the major Z-disc gene product sallimus ( kettin/D-titin , sls [33] , [34] ) as a specific target for HOW in muscle cells and show that sls regulation is dependent on MAPK phosphorylation of HOW . Mechanistically , we demonstrate that HOW phosphorylation is essential for its efficient homodimerization and RNA binding capability . Taken together , our results reveal a molecular mechanism linking muscle-specific MAPK-dependent phosphorylation of HOW to its ability to homodimerize , bind its targets and regulate them , and thereby contribute to muscle sarcomerization .
Examination of HOW revealed two putative MAPK/ERK consensus sites , comprised of Thr followed by Proline ( Pro ) ( TP ) at residues 59 and 64 ( Figure 1A ) . These could also serve as putative sites for other Ser/Thr kinases such as Cyclin Dependent Kinases ( CDKs ) . Importantly , T64 is conserved throughout HOW ( L ) sequences in all annotated Drosophila species , while T59 is found only in closely related species ( Figure 1B ) . In addition to these conserved phosphorylation sites , HOW also contains putative MAPK docking sites . These “D” [35] or “DEF” [36] domains frequently mediate high-affinity interactions between MAPK and its substrates , allowing efficient phosphorylation of the substrate [37] . HOW harbors three such potential D-domain motifs , 85RKQLAA , 121KKEPLTL , and 245KKRQLMELAI , as well as a single DEF domain motif , 151FNF . All four motifs are fully conserved throughout Drosophila species , and partially conserved in other STAR proteins ( Figure 1C ) . The presence of these domains , together with the occurrence of potential phosphorylation sites , led us to test the possibility that HOW is phosphorylated by MAPK/ERK . To start investigating phosphorylation of HOW by MAPK/ERK , we generated a HOW construct in which the putative phosphoacceptor sites T59 and T64 were mutated to Alanine ( HOWTTAA ) , rendering it non-phosphorylatable . Next , we tested whether MAPK/ERK could phosphorylate HOW in vitro . Briefly , HOW ( L ) WT and HOW ( L ) TTAA were transcribed and translated in vitro in the presence of S35-Methionine , incubated with recombinant activated ERK2 , and run on an SDS gel . Incubation of in vitro translated Yan , an established MAPK substrate protein [38] with active Erk2 resulted in a protein mobility shift on SDS Page as compared to the unphosphorylated protein ( Figure 2A , lanes 1–4 ) . Importantly , HOW ( L ) WT but not HOW ( L ) TTAA also displayed a MAPK-dependent mobility shift ( lane 5 , 7 ) , indicating that MAPK phosphorylates HOW and that the phosphorylation event occurs on the predicted residues ( Figure 2A , arrow ) . However , under these conditions , HOW underwent only partial phosphorylation , as a relatively small fraction of HOW was shifted ( see below , Figure 3 ) . We next tested whether HOW is phosphorylated on its MAPK/ERK consensus sites using Drosophila S2R+ cells . HOW ( L ) WT and HOW ( L ) TTAA were expressed in these cells , precipitated with an anti-HOW antibody , and subjected to Western blot analysis using an anti-phospho-Thr-Pro ( anti-pTP ) antibody ( Figure 2B ) . HOW ( L ) WT protein reacted with the anti-pTP antibody , whereas the HOW ( L ) TTAA form did not . This indicated that in unstimulated cells HOW is phosphorylated at least on one TP site . We have also examined the potential phosphorylation of a shorter HOW isoform , HOW ( S ) , which differs from HOW ( L ) only at the C terminus [13] , . Interestingly , although HOW ( S ) WT possesses all of the sites predicted to be phosphorylated by , and to bind MAPK , it did not react with the anti-pTP antibody when tested in a similar manner ( Figure 2B ) . This difference may stem from the distinct subcellular localization of the different HOW isoforms; while HOW ( L ) is nuclear and is thus probably more accessible to active MAPK/ERK , HOW ( S ) is mostly expressed in the cytoplasm . We therefore limited our analysis to the phosphorylation of HOW ( L ) . We further confirmed the identity of the kinase phosphorylating HOW in S2R+ cells by employing the specific MAPKK/MEK inhibitor , U0126 [40] ( Figure 2C ) . Treatment of the cells with U0126 led to a significant decrease in pERK levels ( compare the left and middle lanes in Figure 2C ) . Notably , in the treated cells a marked reduction was observed in the phosphorylation of immunoprecipitated HOW detected by anti-pTP antibody ( top panel ) in comparison to non-treated cells ( phosphorylation was reduced to 0 . 47±0 . 07 fold of control , n = 3 , Figure 2C′ ) . This result indicates that HOW phosphorylation is dependent on MAPKK/MEK . Interestingly , the phosphorylation of HOW on the TP sites was not completely eliminated , as expected from the effective reduction in pERK levels , suggesting that another kinase might also be involved in the phosphorylation of at least one of the TP sites . In a complementary approach , we stimulated MAPK/ERK signaling using Phorbol 12-Myristate 13-Acetate ( TPA/PMA ) . This compound is an activator of protein kinase C [41] , which in turn leads to activation of MAPK/ERK [42] . Treatment with TPA increased HOW phosphorylation , as indicated by stronger reaction with the pTP antibody ( Figure 2D , quantification of 3 experiments is shown in Figure 2D′ ) . Taken together , these experiments demonstrate that HOW is phosphorylated by MAPK/ERK in Drosophila S2R+ cells . What could be the reason for the partial phosphorylation of in vitro translated HOW by activated ERK ( Figure 2A ) ? We considered the possibility that the incomplete phosphorylation is due to a limited degree of HOW dimerization that occurred under our experimental conditions . Accordingly , we examined whether MAPK/ERK phosphorylation was related to the degree of HOW dimerization by performing dimerization experiments with HOWWT and HOWTTAA . Both HOW variants were fused to GFP and expressed in S2R+ cells together with either HA-tagged HOWWT or HOWTTAA , respectively . We immunoprecipitated the GFP-HOW ( L ) protein using an anti-GFP antibody , and tested , using anti-HOW antibody , the co-precipitation of HOW ( L ) -HA ( Figure 3A ) . We differentiated between GFP-HOW , HOW-HA and the endogenous HOW by virtue of their different molecular weights . Indeed , HOWWT-HA readily co-immunoprecipitated with GFPWT-HOW ( Figure 3A , lane1 ) , indicating the two proteins oligomerized , presumably as dimers . A negative control protein ( eGFP ) did not precipitate HOW ( L ) -HA , demonstrating the specificity of the interaction ( Figure 3A , lane 4 ) . Interestingly , the non-phosphorylatable HOWTTAA co-immunoprecipitated less efficiently ( lane 2 ) . Quantification of the experiments , while normalizing the co-precipitated HOW-HA levels to its levels in the crude extract as well as to the immunoprecipitated HOW-GFP , showed that co-precipitation of the phospho-mutant HOW was reduced by about 40% compared to that of HOWWT ( Figure 3A′ ) . In addition , we attempted to examine the effect of enhanced phosphorylation on HOW dimerization , using a putative phospho-mimicking mutant variant , in which the two Thr residues were replaced by Aspartic acid ( HOW ( L ) TTDD ) . However , this mutation did not appear to mimic phosphorylated HOW , in this as well as in other assays ( data not shown ) . As a control for the dimerization assay , we mutated HOW on a Glutamic acid residue , changing it to Glycine ( E106G , HOWEG ) and similarly tagged it with HA or GFP . This residue was previously shown to be essential for QKI [23] and GLD-1 dimerization [24] . As expected , the amount of co-precipitated HOWEG-HA was reduced to about 10% of wild-type levels ( Figure 3A lane 3 and Figure 3A′ ) , confirming the reliability of our dimerization assay . Strikingly , testing the phosphorylation state of HOW dimerization mutant HOWEG using anti-pTP antibody revealed that this HOW mutant essentially does not undergo phosphorylation by MAPK/ERK ( Figure 3B lane 3 , 3B′ ) . Collectively , the above experiments strongly suggest that MAPK-dependent phosphorylation occurs only on HOW dimers , although we cannot exclude a possible change of conformation in the monomeric structure of HOWEG [24] . Our results also further support the idea that phosphorylation stabilizes HOW dimerization since the extent of dimer formation of HOWTTAA is significantly reduced . To address whether the phosphorylation of HOW influences its ability to bind RNA , we performed an RNA binding assay . In this experiment , wild-type HA-tagged HOW ( L ) was affinity purified from S2R+ cells that were either grown in normal medium ( in which HOW is phosphorylated , see Figure 2C ) or in medium containing the MEK inhibitor U0126 ( in which HOW phosphorylation is reduced , shown in Figure 2C ) . The RNA-binding activity of purified HOW-HA was then tested using biotin-labeled RNA oligomers that either contained or did not contain the HOW response element ( HRE ) [43] . A significant reduction of about 70% in the binding of HOW to the RNA was detected following treatment with the MEK inhibitor U0126 ( Figure 4A , left lane ) . In addition , HOW ( L ) EG , the mutant form that does not dimerize , exhibited an extremely low RNA binding activity , supporting a role for the dimerization of HOW in RNA binding ( Figure 4A , right lane ) . Notably , the non-phosphorylatable HOW ( L ) TTAA only showed slightly reduced binding ( Figure 4A , second lane from the right , about 20% reduction ) , possibly due to its ability to dimerize with endogenous , phosphorylated HOW protein ( contrary to the U0126-treated cells , where all HOW proteins are less phosphorylated ) . This experiment demonstrates that ERK-dependent phosphorylation of HOW enhances not only its homodimerization but also its RNA binding activity , suggesting that both functions are linked . Phosphorylation alters the subcellular localization of Sam68 [4] , [32] , raising the possibility that , via this mode of regulation , MAPK impinges on HOW's ability to bind and regulate its RNA targets . To address this issue , we transfected S2R+ cells with HOWWT or HOWTTAA tagged with HA , and stained with an anti-HA antibody . We did not detect any alteration in the subcellular localization of HOW ( L ) TTAA relative to HOW ( L ) WT ( Figure 4B–4C′ ) . Thus , HOW Thr phosphorylation does not alter its subcellular localization . To follow the pattern of HOW phosphorylation in vivo , we generated a polyclonal antibody designed to specifically recognize HOW only when it is phosphorylated on the more conserved Thr residue , T64 ( we refer to the antibody as anti-pHOW ( pT64 ) ( see Materials and Methods ) ) . We first confirmed the specificity of the antibody by transfecting S2R+ cells with how ( l ) WT , how ( l ) TTAA , and by treating a sample of the HOW ( L ) WT expressing cells with the MAPKK/MEK inhibitor U0126 . We performed immunoprecipitation ( IP ) with an anti-HOW antibody and used anti-pHOW ( pT64 ) antibodies in Western blot analysis ( Figure 5A ) . The antibody reacted with the immunoprecipitated HOW ( L ) WT ( lane 2 ) but not with HOW ( L ) TTAA ( lane 4 ) nor with non-transfected cells ( lane 1 ) . In the sample treated with U0126 ( lane 3 ) , the reactivity of the antibody was significantly reduced , suggesting the antibody indeed detects HOW phosphorylation , particularly on T64 . We used the anti-pHOW ( pT64 ) antibody to identify tissues in which HOW is phosphorylated ( Figure 5C ) . To this end , we stained wild-type Drosophila embryos , employing how mutant ( howstru−/− ) embryos , in which zygotic HOW is not produced , as a control for antibody specificity . The embryos were also co-stained with a general anti-HOW antibody . Specific anti-pHOW staining was detected in the nuclei of somatic muscles ( Figure 5C , open arrow ) as well as in the heart cardioblasts ( Figure 5C , white arrow ) . This staining co-localized with anti-HOW staining ( Figure 5D , corresponding arrows ) and was reduced to background levels in howstru mutants ( Figure 5E , 5F ) . Importantly , phosphorylated HOW was confined to the nuclei of the somatic muscles that were marked with muscle-specific expression of CD8-GFP ( which localizes to the cell membrane but often concentrates in the ER surrounding the nuclei ) , as shown in Figure 5G–5I . This result indicates that HOW is phosphorylated on T64 in muscle nuclei . Intriguingly , the anti-pHOW ( pT64 ) antibody did not label tendon cells , which exhibited a strong anti-HOW staining ( Figure 5C , 5D , open arrowheads ) , further validating the specificity of our antibody . The lack of pHOW staining is not simply due to low levels of MAPK in these cells , since previous analysis demonstrated high level of phospho-ERK in tendon cells at this stage [44] . We therefore hypothesized that the HOW ( L ) isoform found to be specifically phosphorylated by MAPK/ERK ( Figure 2 ) is not expressed at high levels in tendon cells at this developmental stage and that the anti-HOW staining is mainly detecting HOW ( S ) , which does not undergo phosphorylation in S2R+ cells ( Figure 2B ) . To directly address this possibility , we generated transgenic flies expressing HA-tagged versions of HOW ( L ) WT or HOW ( L ) TTAA under UAS-GAL4 binding sequences . The expression of HOW ( L ) WT or HOW ( L ) TTAA was driven in embryonic tendon cells by the tendon-specific driver , sr-GAL4 , and the embryos were stained for pHOW ( Figure 5J–5M ) Whereas over expression of HOW ( L ) WT in tendon cells led to positive nuclear staining with the anti-pHOW antibody ( Figure 5J , 5K arrows ) , overexpression of HOW ( L ) TTAA did not result in such a staining ( Figure 5L , 5M arrows ) . This experiment shows that the kinase required for HOW phosphorylation on T64 is activated in tendon cells , and that the lack of reactivity of the antibody in wild-type tendon cells at late embryonic stages is due to low levels of HOW ( L ) in this tissue . This is also consistent with the cytoplasmic HOW staining characteristic of the HOW ( S ) isoform present at stage 16 embryos [39] . Based on these results , we conclude that HOW ( L ) is phosphorylated on T64 in the nuclei of somatic muscles and heart cardioblasts . To further verify that HOW undergoes Thr phosphorylation in embryonic somatic muscles , we also expressed HA–tagged HOW ( L ) WT or HOW ( L ) TTAA in muscles using the mef2-GAL4 driver . The HOW variants were subsequently immunoprecipitated using anti-HA conjugated beads and reacted with anti-pHOW ( pT64 ) antibodies on Western blots ( Figure 5B , lane 2 ) . Only HOW ( L ) WT but not HOW ( L ) TTAA ( lane 3 ) , was detectable by the anti-pHOW antibody . In addition , an anti-HOW antibody identified an additional upper band that might represent phosphorylated HOW ( lane 2 ) . To conclude , these findings strongly indicate that HOW ( L ) undergoes phosphorylation on T64 in embryonic somatic and cardiac muscles and that phosphorylated HOW in muscles is localized specifically to the nucleus . To further characterize the physiological significance of MAPK/ERK-dependent phosphorylation of HOW , we focused on somatic muscles , a tissue in which a high degree of HOW phosphorylation was detected . Recently , HOW was shown to be essential for muscle sarcomerization [45] . Specifically , in larvae where how was knocked down ( using how RNAi expressed specifically in muscles ) , the sarcomeric organization is aberrant . In these larvae , the Z discs appear discontinuous and spotty , a phenotype similar to that caused by the knock down of several genes encoding sarcomeric proteins [45] . To address the possibility that this phenotypic resemblance is a result of regulation of one or more of these proteins by HOW , we assessed the levels of different sarcomeric proteins in 3rd instar larvae , in which we reduced HOW levels using RNAi mediated knock-down in muscles . To enhance the effect of how down-regulation , the larvae used were also heterozygous for howstru . Under these conditions , we observed an increase in the 500 kDa isoform of Sls , a giant protein that serves as a scaffold in the sarcomere , linking the Z-discs to the thick filaments [33] ( Figure 6A ) . In contrast , the levels of MSP-300 ( isoform 800 kDa ) , a Z-disc protein [46] , [47] , were decreased ( Figure 6A ) . This suggests that in wild-type larval muscles , Sls is down-regulated by HOW , while MSP-300 is elevated . Although these results do not distinguish between direct and indirect effects , nonetheless , we used the levels of these two proteins in muscles as readout of the activity of phosphorylated ( HOWWT ) versus non-phosphorylated HOW ( HOWTTAA ) in this developmental process . Accordingly , we expressed comparable amounts of either HOWWT or HOWTTAA in muscles ( using the mef2-GAL4 driver ) and followed the expression of Sls and MSP-300 by fluorescent labeling ( Figure 6B–6D ) or by Western blot analysis ( Figure 6E ) in 3rd instar larvae . As expected , over-expression of HOW ( L ) WT resulted in reduced levels of Sls ( Figure 6C ) . This was confirmed by measuring the immunofluorescence intensity of Sls in Z-discs ( see legend to Figure 6 ) . Importantly , overexpression of HOW ( L ) TTAA exhibited a significantly milder effect on Sls levels ( Figure 6D ) . In contrast , over-expression of either HOWWT or HOWTTAA resulted in a similar aberrant distribution of MSP-300 ( Figure 6C′ , 6D′ ) . We also quantified the changes in protein levels by employing Western blot analysis of extracts from the body walls of several larvae , expressing driver alone ( mef2-GAL4 ) , or together with HOWWT or HOWTTAA ( Figure 6E ) . Consistent with the immunofluorescent staining , the anti-Sls antibody reacted with a major band of 500 kDa that was downregulated following HOW ( L ) WT overexpression . A much milder effect was observed following HOW ( L ) TTAA overexpression ( Figure 6E , upper band ) . Myosin Heavy Chain ( MHC ) [48] , [49] was reduced following expression of both HOWWT and HOWTTAA ( Figure 6E second band from top ) . Of the three distinct bands representing MSP-300 ( 900 kDa , 800 kDa , 300 kDa ) , the 900 kDa and the 300 kDa bands were elevated , whereas that of 800 kDa was unaffected following HOWWT overexpression . No significant differences were found between the effect of HOWWT and HOWTTAA on all three bands . The levels of two additional muscle proteins , MLP84B [46] and Actin remained unchanged in these genetic backgrounds . To conclude , the down-regulation of the newly-identified HOW target , Sls , is dependent on the phosphorylation state of HOW , whereas MSP-300 and MHC are regulated by HOW in a phosphorylation-independent manner . This discrepancy may be due to differential activities of HOW ( e . g . regulation at the level of RNA degradation versus alternative splicing ) . Given that HOW is an RBP , we next sought to determine whether HOW controls sls RNA levels , and if so , whether phosphorylation is important for this type of regulation . To this end , we purified total RNA from the body walls of single 3rd instar larvae and performed real-time PCR to quantify the levels of the sls and how transcripts ( rp49 RNA served as control ) using the SYBR green method . We compared larvae from a wild-type background ( w− ) , to larvae heterozygous for howstru which expressed how RNAi in muscles , and to larvae over-expressing either HOW ( L ) WT or HOW ( L ) TTAA in muscles . In larvae expressing lower levels of how ( Figure 6F′ , second left bar , RNAi ) , sls mRNA was upregulated ( Figure 6F ) . Importantly , HOW ( L ) WT was more efficient in the down-regulation of sls RNA relative to HOW ( L ) TTAA ( Figure 6F right bars ) . Although this result was not statistically significant , the trend seen at the level of RNA is consistent with the results obtained in the protein analysis . Thus , our data demonstrate that HOW ( L ) has a role in the down-regulation of sls mRNA and protein levels in muscles , and that this process is dependent on HOW phosphorylation . We next examined whether HOW phosphorylation in muscles , and its resulting elevated activity in the regulation of Sls levels ( Figure 6 ) , are indeed dependent on MAPK signaling . To test this idea , we expressed RNAi for the Drosophila MAPK gene rolled in larval muscles under mef2-GAL4 regulation , immunoprecipitated HOW from protein extracts of these larvae and examined HOW phosphorylation using Western blot analysis with an anti-pTP antibody ( Figure 7A ) . Importantly , in the larvae where rolled was down regulated ( note the reduction in the levels of pERK in Figure 7B ) , we observed diminished phosphorylation of HOW , compared to wild-type larvae ( Figure 7A ) . The phosphorylation of HOW in wild-type larvae was present in a band corresponding to the HOW ( L ) protein . Thus , phosphorylation of HOW ( L ) in larval muscles in vivo is dependent on MAPK signaling , as in S2R+ cells and in an in vitro kinase assay ( Figure 2 ) . Since rolled down regulation resulted in decreased HOW ( L ) phosphorylation , we hypothesized that HOW ( L ) activity will be lowered in these larvae , resulting in elevated levels of Sls . Indeed , using a Western blot analysis ( Figure 7B ) we find that in larvae expressing rolled RNAi under mef2-GAL4 , Sls levels are significantly elevated . The increase in Sls levels is specific , as only a mild increase is observed in MHC levels , while MSP and α-Actinin levels are unaffected ( Figure 7B ) . The marked elevation of Sls levels following reduction of rolled in the larval muscles is also readily detectable in individual muscles ( Figure 7C , 7F ) , while MSP-300 levels remain constant ( Figure 7D , 7G , merge in Figure 7E , 7H ) . Thus , we conclude that in wild-type larval muscles , MAPK signaling is required for HOW ( L ) phosphorylation and reduction of Sls protein levels .
STAR proteins regulate tissue differentiation in a wide range of species including nematodes [50] , flies [17] , Zebrafish [8] , mice , and humans [51] . They function by controlling diverse posttranscriptional events , often forming specific protein-RNA complexes mediated by 3′UTR sequences of their target mRNAs , or with alternatively spliced introns [2] . In this study we reveal a molecular mechanism regulating the activity of the STAR protein HOW . Our findings demonstrate that HOW is phosphorylated on Thr residues embedded within conserved MAPK consensus sequences , both in cultured cells as well as in muscle cells and heart cardioblasts ( Figure 2 , 5 ) , and that this phosphorylation is executed by MAPK/ERK ( Figure 2 , 7 ) . Significantly , our results provide novel molecular insights to the importance of this phosphorylation , demonstrating that phosphorylation stabilizes HOW dimer formation ( Figure 3 ) . Moreover , because phosphorylation presumably occurs only on dimers , and phospho-dimers are more stable , we propose that a feed-forward loop ensures that a large fraction of HOW dimers are formed following a short temporal burst of MAPK/ERK activation . Importantly , since STAR proteins are evolutionarily conserved , the novel mode of regulation that we have uncovered might have important implications for other members of the QKI sub-family . Two of the four potential MAPK/ERK docking sites reside within the HOW QUA1 domain . The close proximity between these sites , the MAPK/ERK phosphorylation sites , and the E106 residue critical for dimerization ( [23] , [24] , Figure 3 ) , all of which are highly evolutionarily conserved ( Figure 1 ) , is consistent with local conformational changes in the QUA1 domain induced by phosphorylation , which further lead to dimer formation and stabilization . HOW dimerization might be essential for a number of its characteristics; first , it enables binding to several HREs present on a single target RNA , leading to an overall higher affinity of HOW to its target RNA . For example , the 3′UTR of the HOW target stripe contains three consecutive HREs and a half site that resides between the first and second HREs . We have previously shown that a higher affinity of HOW is observed when multiple HREs are clustered together [43] . HOW dimerization and binding to two sites may also contribute to the formation of a secondary RNA structure , thus facilitating RNA processing . Interestingly , HOW dimerization apparently enhances its RNA binding ability even to a single binding site ( Figure 4A ) . Additionally , HOW dimerization might also potentiate the recruitment of other proteins/enzymes to the vicinity of the targeted mRNA . This may occur either if one subunit associates with the target RNA , and the other with a specific enzyme that induces modification/degradation of the target RNA , or if the formation of HOW dimers enables recruitment of proteins that would not associate with its monomeric form . A similar mechanism has been proposed to take place when GLD-1 dimers associate with their RNA target ( s ) close to the polyadenylation site , possibly recruiting an E3 ubiquitin ligase complex to this site . Ubiquitination of the polyadenylation complex would subsequently lead to shortening of the poly-A tail [24] . A mechanism for HOW activity that includes protein interactions with binding partners has yet to be described , but such an interaction is highly likely , since HOW does not contain any recognizable catalytic domains that could independently lead to mRNA degradation . We show that a HOW mutant unable to form dimers exhibits a greatly diminished RNA binding capacity ( Figure 4A , HOWEG mutant ) . This is also in line with our findings that HOW binding to RNA is significantly reduced when it is hypophosphorylated , and that phosphorylated HOW has a higher tendency to form dimers . We conclude that dimeric HOW binds RNA with higher affinity than monomeric HOW . This conclusion is in line with the observations that GLD-1 lacking its QUA1 domain has a lower affinity to RNA [26] , and possibly applies to other STAR proteins . To our knowledge , this is the first example of a positive effect of phosphorylation of a STAR protein on its RNA binding capacity . Phosphorylation of Sam68 by MAPK/ERK was demonstrated to have a subtle negative influence on its RNA binding [32] , [52] , while Tyrosine phosphorylation was shown to more severely impair RNA binding of both QKI [29] and Sam68 [53] . As the Tyrosine residues in HOW are highly conserved with those of QKI , it is highly likely that they are also phosphorylated ( Kirenberg and Volk , unpublished data ) , and that this phosphorylation may have an opposite , negative effect on the ability of HOW to bind its targets . Hence , it is interesting to note that changes in cellular signaling may have the capacity to fine-tune , both positively and negatively , the ability of an RNA binding protein to bind its targets , thus modulating the levels of a variety of mRNAs . A requirement for HOW in developing muscles had been demonstrated previously [45] , but its target RNAs in this tissue have not been characterized . In this study , we identify three different muscle proteins , Sls , MSP-300 and MHC , whose levels are altered by the expression of HOW in this developmental setting . We still do not know whether the mRNAs of these proteins are all directly bound by HOW , and at what level the regulation by HOW occurs , i . e . via control of specific alternative splicing or by regulation of overall mRNA levels . Since sls mRNA levels respond to both HOW overexpression and knock-down , and given that the sls transcript contains several potential binding sites for HOW ( both at the 3′UTR as well as within several introns; not shown ) , it likely represents a true direct RNA target of HOW . Even though Sls is a structural protein , its fast turnover in sarcomeres might be essential for maintenance of the sarcomeric architecture [54] . To fulfill this requirement , its protein and RNA half-life should be short . Thus , HOW might play an essential role in promoting destabilization of sls mRNA to promote fast exchange of newly formed Sls protein at the Z-disc . Indeed , in muscles where MAPK signaling was downregulated using rolled RNAi , a significant elevation in Sls levels is clearly evident ( Figure 7 ) . This result is in line with tight regulation of Sls levels occurring in wild-type larvae . Although the precise RTK signaling pathway that regulates HOW phosphorylation in muscles is yet to be elucidated , the FGFR Heartless , which is expressed by muscle cells throughout their development [55] , [56] , is an attractive candidate . It is possible that continuous FGFR activation in muscles promotes HOW phosphorylation by MAPK , rendering it more active in controlling the levels of its target mRNAs in this tissue . In summary , in this study we have unraveled and characterized a novel molecular mechanism at the basis of the activity of the STAR protein , HOW . By linking its activity to MAPK/ERK-dependent phosphorylation and regulation , we provide a mechanistic linkage between HOW phosphorylation , the degree of its dimerization and its biological activity/function . We propose that this mechanism may apply to other STAR proteins , in which the dimerization domain and phosphorylation sites are evolutionarily conserved .
Mutant HOW ( L ) constructs ( TTAA , EG ) were created by site directed mutagenesis ( Stratagene ) , following the manufacturer's protocols . pUAST-HOW ( L ) constructs tagged with HA or fused with GFP were generated in the pTWH and pTGW Drosophila Gateway vectors , respectively ( T . Murphy , Carnegie Institution of Washington ) using the Gateway cloning system ( Invitrogen ) . Fly stocks used in this study include w− , mef2-GAL4 ( Bloomington stock center ) , mef2-GAL4 , UAS-CD8 GFP ( F . Schnorrer , Martinsried , Germany ) , stripe-GAL4 ( G . Morata , Madrid , Spain ) , howstru [57] , how RNAi line ( ds-HOW ) [43] , rolled RNAi ( TRiP HMS00173 , Bloomington stock 34855 ) . UAS-how ( l ) WT-3HA and TTAA-3HA were injected to flies by Genetic Services , Sudbury , MA , USA . Primary antibodies used in this study include mouse anti-phospho-Thr-Pro ( p-Thr-Pro-101 ) ( Cell Signaling Technology ) , rabbit and rat anti-HOW [39] , guinea pig anti-MSP-300 [47] , rat anti-Sls ( Kettin ) ( Klg16 , MAC155 , Abcam ) , rat anti α-Actinin ( Abcam ) , rabbit anti-MHC ( P . Fisher , Stony Brook , NY ) , mouse anti-Actin ( Sigma ) , rabbit anti-Mlp84bB [58] , mouse anti-GFP ( Roche ) , mouse anti-HA ( Roche ) , chick anti-HA ( Aves Labs ) , mouse anti-Lamin ( [59] , gift from Y . Gruenbaum , Hebrew University ) , mouse anti-pERK ( gift from B . Shilo , Weizmann Institute ) . The polyclonal anti-pHOW ( pT64 ) antibody was raised in rats against the phospho-peptide PQHL ( p ) TPQQ ( generated by Sigma ) , corresponding to amino acids 60–67 of HOW . Immunizations were performed by the antibody unit at the Weizmann Institute . The serum was cleaned on beads conjugated to a similar peptide , PQHLQPQQ , in an attempt to reduce background staining that was suspected to be due to another protein containing this sequence . Secondary antibodies used in this study include various Cy3 , Cy2 , Cy5 and HRP-conjugated antibodies ( Jackson ImmunoResearch Laboratories , USA ) . S2R+ cells were grown and transfected essentially as previously described [43] , except that cells were usually collected for analysis 36 hours after transfection . For inhibition of MAPK/ERK activity , 10 µM U0126 ( Sigma ) in DMSO ( or only DMSO for controls ) was administered with the serum-containing media , about 18 h after the transfection . Since U0126 led to a decrease in the efficiency of recovery from transfection , the control cells were transfected with smaller amounts of how DNA . Phorbol 12-Myristate 13-Acetate ( TPA/PMA ) ( Sigma ) treatment ( 10 µM in DMSO , or only DMSO for controls ) was performed on starved cells , 20 h after transfection , for 15 min . Cells were scraped off the flasks and collected in PBS , washed twice and lysed in 1% NP40 lysis buffer ( 50 mM Tris pH 7 . 5 , 1% NP40 , 100 mM NaCl , 1% Protease inhibitor cocktail ( P8340 , Sigma ) , 0 . 5% Phosphatase Inhibitor Cocktail 1 ( P2850 , Sigma ) , 20 mM β -glycerol phosphate ) . Embryos and larvae were collected and crushed in RIPA buffer ( 1% sodium deoxycholate , 1% Triton X-100 , 0 . 1% SDS , 0 . 15M NaCl , 0 . 05M Tris pH 7 . 0 , supplemented with the same inhibitors ) . For IP , equal amounts of protein lysate were incubated with protein A/G beads ( SC-2003 , Santa Cruz ) coupled with either rabbit anti-HOW polyclonal antibody or mouse anti-GFP monoclonal antibody , or agarose HA conjugated beads ( A2095 , Sigma ) for 1–2 h at 4°C . Beads were washed three times with the NP40 lysis buffer , and boiled in protein sample buffer to elute the proteins . MAPK/ERK in vitro phosphorylation assay was performed as described [60] . High molecular weight proteins were analyzed by SDS–PAGE using 2 . 5% acrylamide gels strengthened with 1 . 5% agarose , essentially as in [61] . Fixation and staining of embryos were done following standard procedures . Larval Flat Preparations were performed essentially as described [62] , except the fixation was done in 4% PFA , the primary antibody staining was carried out for 2 h and the secondary for 1 h . Quantification of average intensity of larval muscles images were performed using MATLAB . S2R+ cells were seeded on Ibidi u-Slide 8 well ( 0 . 2*106 cells per well ) , fixed with 3% PFA for 5 min , permeabilized using 3% PFA+0 . 1% TritonX-100 , stained with primary antibody ( 1∶200 ) for 30 min , and secondary antibody ( 1∶400 ) for 30 min . Visualization was carried out using a Zeiss LSM710 confocal system . Single larvae were flipped over and their interior was cleaned , leaving the carcass only . RNA was extracted by the Nucleospin Purification Kit ( Macherey-Nagel ) according to the manufacturer's instructions . Equal amounts were used as a template for cDNA preparation using the Verso cDNA kit ( Thermo Scientific ) . Real time PCR was carried out using Fast SYBR Green Mastermix ( Applied Biosystems ) in a StepOne plus machine ( Applied Biosystems ) . The following primers were used: sls ( TGCCCATGCCGAAGACA ) and ( TGTCTTGTTTGCTGTTACGTTTACAG ) , rp49 ( GACCATCCGCCCAGCATAC ) and ( CCATTTGTGCGACAGCTTAGC ) , how ( AACTTTGTCGGTCGCATTTT ) and ( CGTCCTCCTTCTTCTTGTCG ) . In vitro RNA–protein binding assay was performed essentially as described [43] , only that the proteins were not in vitro translated , but were purified from S2R+ cells using immunoprecipitation with HA conjugated beads , followed by elution with an HA peptide ( Sigma , I2149 ) . Phosphorylation sites were predicted by the GPS2 . 1 program [63] . Protein scheme was generated by DOG2 . 0 software [64] . Student's t-tests were performed using GraphPad software . | Somatic muscles are huge cells that feature highly organized sarcomeric architecture , whose formation and maintenance are not fully understood . Multiple signals play a role in these cells , including the highly conserved MAPK/ERK pathway , which often serves as a cue for cellular proliferation or differentiation . In this study , we reveal a role for MAPK/ERK signaling in the regulation of multiple muscle genes at the level of mRNA , through the control of the activity of the Drosophila RNA–binding protein Held Out wings ( HOW ) . Specifically , we show that HOW undergoes phosphorylation by MAPK/ERK , which increases its ability to form dimers and enhances its RNA–binding capacity . We further demonstrate that HOW is phosphorylated in embryonic and larval muscles by MAPK in vivo and that this event is important for its ability to regulate the levels of a giant sarcomeric gene homologous to vertebrate titin , thus contributing to the maintenance of muscle sarcomeric architecture . Importantly , HOW is a close homolog of mammalian Quaking , an essential protein in embryonic development and nervous system myelination , a reduction of which is correlated with Schizophrenia . Thus , our results raise the possibility that MAPK/ERK phosphorylation similarly regulates RNA profiles in other tissues controlled by proteins of the conserved Quaking family . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"anatomy",
"and",
"physiology",
"signaling",
"in",
"selected",
"disciplines",
"muscle",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"molecular",
"development",
"musculoskeletal",
"system",
"signaling",
"in",
"cellular",
"processes",
"gene",
"expression",
"biology",
"molecular",
"biology",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"nucleic",
"acids",
"physiology",
"genetics",
"cellular",
"types",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"signaling",
"cascades"
] | 2012 | Phosphorylation of the Drosophila melanogaster RNA–Binding Protein HOW by MAPK/ERK Enhances Its Dimerization and Activity |
Cochlear outer hair cells ( OHCs ) are fast biological motors that serve to enhance the vibration of the organ of Corti and increase the sensitivity of the inner ear to sound . Exactly how OHCs produce useful mechanical power at auditory frequencies , given their intrinsic biophysical properties , has been a subject of considerable debate . To address this we formulated a mathematical model of the OHC based on first principles and analyzed the power conversion efficiency in the frequency domain . The model includes a mixture-composite constitutive model of the active lateral wall and spatially distributed electro-mechanical fields . The analysis predicts that: 1 ) the peak power efficiency is likely to be tuned to a specific frequency , dependent upon OHC length , and this tuning may contribute to the place principle and frequency selectivity in the cochlea; 2 ) the OHC power output can be detuned and attenuated by increasing the basal conductance of the cell , a parameter likely controlled by the brain via the efferent system; and 3 ) power output efficiency is limited by mechanical properties of the load , thus suggesting that impedance of the organ of Corti may be matched regionally to the OHC . The high power efficiency , tuning , and efferent control of outer hair cells are the direct result of biophysical properties of the cells , thus providing the physical basis for the remarkable sensitivity and selectivity of hearing .
Outer hair cells ( OHC ) in the mammalian cochlea are essential to the remarkable sensitivity of hearing . These highly specialized cells actively feed mechanical power into the organ of Corti and amplify its mechanical vibrations in response to sound [1]–[5] . How this is achieved at auditory frequencies is a subject of considerable debate . Five biological motor mechanisms have been described in outer hair cells that may contribute [2] , [3] , [5] , [6] . Motors localized to the hair bundles include: actin-myosin motors associated with slow bundle movements and adaptation mechano-electrical transduction ( MET ) currents [7] , [8]; Ca2+ sensitive reclosure or conformational change of the MET molecular apparatus associated with fast bundle movements and adaptation [9]; and electrically-driven bundle displacement that act independent of MET function [10] . Motors localized to the soma include: cytoskeletal remodeling mechanisms [11] , [12] and electrically-driven changes in length [13]–[15] . The ability of each of these mechanisms to feed mechanical power into cochlea is limited by their intrinsic thermodynamic properties . As such , some of these motors can be ruled out as key to amplification of mechanical motions in the cochlea simply because they are too slow . The mammalian cochlear amplifier is extremely fast and capable of cycle-by-cycle action , in some species at frequencies exceeding 50 kHz [16] , [17] . This rules out mechanisms that require cyclic phosphorylation , transport and/or protein synthesis . In non-mammalian species , that do not have OHCs or the protein prestin , bundle-based motors underlie the active amplification process [18] , [19] . In mammals , the evidence indicates OHC somatic motility is a key contributor [20]–[24] , and this is the motor we focus on here . OHC somatic electromotility is driven by the MET current entering the cell and likely draws thermodynamic power from the electo-chemical potential between fluid compartments in the cochlea . The apical surfaces of OHCs are bathed in high-potassium endolymph , biased to approximately +50 to +80 mV , and their basal poles bathed in high-sodium perilymph at 0 mV reference . This endocochlear potential is maintained by the stria vacularis and associated cells [25]–[27] . When the hair bundle is displaced and MET channels open at the tips the stereocilia , ionic currents ( primarily K+ and Ca2+ ) are driven into the OHC . A fraction of this MET current enters the apical face of the soma at the base of the stereocilia . In the absence of phosphorylation , it is likely that this current carries the thermodynamic electrical power input that drives the OHC mechanical power output . Here , we analyze how this electrochemical energy is converted into useful mechanical work by somatic electromotility using the model illustrated in Fig . 1 . The current model is fundamentally piezoelectric in nature and extends concepts developed by Iwasa [28] , [29] to address frequency-dependent power conversion efficiency . New results include the force vs . velocity , and power vs . velocity curves for OHCs ( c . f . skeletal muscle cells [30] ) , and the frequency-dependent power efficiency that arises from intrinsic limitations on impedance matching between the cell and the load . Results indicate that OHCs are broadly tuned to have maximum power efficiency at a best frequency , thus contributing to tuning and the place principle in the cochlea . Furthermore , results provide an interpretation of how efferent activation may directly attenuate and de-tune the power output of OHCs and thereby providing a means for the brain to command exquisite control over the cochlear amplifier in a frequency dependent manner .
Dissipative drag from the cytoplasm and the extracellular space are unavoidable . As a first approximation we modeled the axial component of the drag acting on the plasma membrane using a version of the Navier-Stokes equations . Assuming small displacements from the resting configuration , and ignoring the convective nonlinearity , the Navier-Stokes equations reduce to ( 20 ) where is the density of the fluid , r is radial coordinate , is the effective viscosity , and is the axial velocity . To approximate the visco-elastic properties of the materials , we used a complex-valued viscosity of the form , where , is the frequency , is a material constant , and the is a parameter that determines the relative contributions of viscosity vs . elasticity of the material . When this model reduces to the standard Newtonian viscous fluid and when this reduces to the standard shear elastic solid . For biological materials ζ falls between these two extremes – e . g . for the tectorial membrane [55] . These equations account for both the visco-elastic drag and entrained fluid mass . We solved the equations to obtain the velocity field resulting when a cylinder oscillates in the axial direction with displacement . Having the velocity field , we computed the axial shear stress acting on the cylinder wall per unit axial displacement ( 21 ) where are Hankel functions , is the non-dimensional Womersley number ( complex-valued ) , and a is the cylinder radius . With this , the damping parameter appearing in the momentum equation ( Eq . 8 ) is . This model is approximate , but matches the viscous analysis of Tolomeo and Steel [43] if the length of the cell is much longer than the diameter , motions are axial , and the viscosity is strictly real valued , i . e . . Model parameters were estimated from known dimensions and physical constants combined with voltage clamp and mechanical data shown in Figs . 2–3 as well as microchamber data in Fig . 4 . All other results ( Fig . 5–8 ) and voltage clamp data in Fig . 4 are model predictions and the associated data were not used to estimate parameters . The model uses a reference thickness to describe the multi-component composite lateral wall and it is important to note that some parameters cannot be independently separated from this reference thickness ( e . g . , , appear as groups ) . Coefficients appearing in the cable equation were computed from the physical parameters listed below using: , , and . Coefficients appearing in the wave equation were computed using , and . Dimensions were based on OHCs from the guinea pig cochlea . Data in Fig . 2–3 were used to find the effective stiffness , piezoelectric coefficient , electrical permittivity and conductance of the membrane . These data are for relatively low stimulus frequencies where the intracellular axial resistance has negligible effect on the results . To estimate the axial resistance we used the corner frequency where the capacitance measured at the basal pole of the cell begins to roll off ( Fig . 3 ) . The fraction of the membrane occupied by the motor was set to 80% ( ) and the passive component to 20% ( ) . The overall cell compliance was estimated from the slope of the compliance vs . cell length reported by Frank et al . [48] , reproduced in Fig . 2C , using as well as the gain reproduced in Fig . 4 ( solid , microchamber curve ) . An iterative optimization routine was run to refine the initial estimates of , and to simultaneously fit data in Fig . 2–4 . Specific optimized numerical parameters include: OHC radius a = 4 . 5e-6 m; composite mechanical stiffness C* = 1 . 4e6 N/m2 ( based on , and ) ; plasma membrane conductance ; apical face membrane conductance ; basal membrane conductance ; transduction current gain ; composite reference thickness ; OHC length ; length of the active lateral wall was , and was set by requiring passive basal pole to have a passive capacitance of 7 pF; intracellular axial resistance ri = 5 . 76e10 Ohm/m; composite piezoelectric coefficient at rest ( C/m2 ) at rest; plasma membrane area specific capacitance ; density ; transduction current adaptation time constant ; fluid viscosity ; and fractional viscosity coefficient . We note that the mixture fraction is not uniquely determined by currently available data and it is possible to find alternative mixture fractions and stiffness parameters that result in the same composite stiffness C* . Nevertheless , it was necessary to use a value of to simultaneously fit all of the data and explain the magnitude of voltage dependent capacitance under unloaded and zero strain conditions . Additional experiments , perhaps involving voltage-dependent capacitance measurements under controlled mechanical loads , have the potential to resolve this ambiguity and reveal more about the lateral wall motor , but are not necessary for the purpose of the present power analysis since the composite parameters would not change . Experimental procedures and animal care were designed to advance animal welfare and were approved by the Baylor College of Medicine animal care and use committee . All physical parameters were deduced from the published literature , with the exception of the intracellular electrical resistance , . To estimate , we isolated OHCs from the guinea pig cochlea [56] and examined the frequency dependence of the input electrical impedance under whole-cell voltage clamp ( Axopatch 200 B , Molecular Devices , Sunnyvale , CA ) . OHCs were harvested from euthanized guinea pigs . Cells were patch-clamped at the base with quartz pipettes covered with Sylgard , and hyperpolarized to minimize the voltage-dependent nonlinear capacitance . K+ and Ca2+ ion channels were blocked with the addition of ( C2H5 ) 4N ( Cl ) , CsCl and CoCl to the bathing and/or pipette solutions [57] . The input admittance was determined with a single sinusoidal voltage ( 0 . 015 V peak to peak , 90–3200 Hz ) superimposed on top of a −0 . 13 V holding potential after correcting for the inherent phase shifts of the amplifier [58] . 210 measurements were averaged at each frequency . The resistance and capacitance were calculated from the input admittance [57] accounting for the series resistance ( ∼6 Mohm , remained constant throughout experiment ) . Experiments were conducted at room temperature .
There are four major observations that can be drawn from the present work . The first addresses how OHCs operate at high frequencies given their electrical capacitance [9] , [62] , [81]–[83] . Capacitance is thermodynamically conservative and present results confirm that the ability of OHCs to supply mechanical power to the cochlea is not limited by electrical capacitance [84] , even at frequencies much higher than the membrane time constant ( e . g . Fig . 7 ) . This is true because capacitance is not dissipative . Instead , present results suggest the most serious factor that may limit power output by OHCs is how well the “impedance” of the hair cell is matched to that of the cochlear partition ( e . g . Fig . 6 ) . OHCs driving against an excessively stiff cochlear partition , for example , would be inefficient . The second observation is that OHCs may be tuned to maximize their power output at a best frequency , albeit broadly tuned . Although OHC displacement and force are quite flat over a broad range of frequencies when driven by voltage ( e . g . Fig . 5 , present model and published data [48] ) , OHC power output is tuned when one considers the mechanical power output relative to the electrical power input . The predicted tuning is dependent upon cell length and correlates with the cochlear place principle [78] , thus indicating that tuning of OHCs may contribute to the sharp mechanical and afferent neural tuning in the living cochlea . The third observation addresses how the MET channels would be expected to further tune output of the somatic motor . MET adaptation generates high-pass filtered MET currents [47] , [66] , [85] , [86] . Since the filtering is upstream of the somatic motor it would further sharpen tuning of OHC somatic motor output by attenuating low-frequency amplification . In the context of the organ of Corti , MET adaptation would also be expected to alter the phase of the OHC force possibly to maximize power input to the cochlea near the best frequency [67] and , additionally , might introduce a non-optimal phase that would sharply attenuate cochlear gain at both low and high frequencies . Because of these factors , the influence of tuning in isolated OHCs on tuning curves in the cochlea would be expected to be even more significant than implied by the OHC motor efficiency alone ( Fig . 7 ) . The fourth observation is that OHC somatic power output may be controlled by the brain via efferent activated ionic conductance ( s ) . The model predicts that increasing the conductance of the basal pole would reduce OHC power output and tuning , thus providing a plausible explanation for a fast mechanism that may be used by the brain to control both sensitivity and frequency selectivity of hearing ( e . g . Fig . 7 ) . Finally , it is important to note that the OHC somatic motor is not present in non-mammals , yet these animals also exhibit many of the properties of the mammalian cochlear amplifier [87] , [88] . The MET apparatus itself is clearly a key contributor to hair bundle motility and amplification [9] , [89] . In addition , there is an MET-independent component of hair bundle motility driven by voltage [10] . This voltage-dependent component has analogy to the somatic motility addressed here , and may be involved in tuning and the power stroke of hair bundle motility with potential relevance to active bundle amplification in high frequency hearing organs [90] . These hair-bundle features occur upstream of the somatic motor and the two clearly interact with each other via micromechanical environment and electrical fields [91] . | The sense of hearing is exquisitely sensitive to quiet sounds due to active mechanical amplification of sound-induced vibrations by hair cells within the inner ear . In mammals , the amplification is due to the motor action of “outer hair cells” that feed mechanical power into the cochlea . How outer hair cells are able to amplify vibrations at auditory frequencies has been somewhat of a paradox given their relatively large size and leaky electrical properties . In the present work , we examined the power conversion efficiency of outer hair cells based on first principles of physics . Results show that the motor is highly efficient over a broad range of auditory frequencies . Results also show that the motor is likely controlled by the brain in a way that allows the listener to focus attention on specific frequencies , thus improving the ability to distinguish sounds of interest in a noisy environment . | [
"Abstract",
"Introduction",
"Methods",
"Discussion"
] | [
"physiology/sensory",
"systems",
"neuroscience/sensory",
"systems",
"computational",
"biology"
] | 2009 | Power Efficiency of Outer Hair Cell Somatic Electromotility |
Tapeworms of the order Diphyllobothriidea are parasites of tetrapods and several species may infect man and cause neglected human disease called diphyllobothriosis . Identification of human-infecting diphyllobothriid cestodes is difficult because of their morphological uniformity , which concerns also their eggs in stool samples . In the present study , we analysed by far the largest dataset of more than 2 , 000 eggs of 8 species of diphyllobothriid cestodes that may infect humans , including the most frequent human parasites Diphyllobothrium latum , D . nihonkaiense and Adenocephalus pacificus ( syn . Diphyllobothrium pacificum ) . Size ( length , width and length/width ratio ) and the surface of the egg shell from naturally and experimentally infected hosts were studied using light and scanning electron microscopy . A high degree of intraspecific and host-related size variability has been detected , but combination of morphometrical and ultrastructural data made it possible to distinguish all of the studied species , including otherwise quite similar eggs of the 3 most common species infecting man , i . e . D . latum , D . nihonkaiense and D . dendriticum . The surface of all marine species is covered by numerous deep pits with species-specific density , whereas the surface of freshwater species is smooth or with isolated shallow hollows or wrinkles .
Human infections with gastrointestinal helminths are usually diagnosed when their eggs are found in stool samples . Therefore , examination of stool and detection of parasite eggs is used as a non-invasive method for detection of infections [1] . Some of the helminth parasites are relatively easy to identify to the species level based on egg morphology , such as species of Schistosoma Weinland , 1858 ( blood flukes ) [1] . In contrast , eggs of diphyllobothriid tapeworms including human-infecting species are highly variable , their size may overlap between species and is also influenced by individual definitive hosts and their size . This makes reliable species diagnosis of causative agents of human diphyllobothriosis in stool samples difficult or even impossible [2] . Diphyllobothriosis and closely related diplogonoporosis are fish-borne zoonoses that are neglected , but widely distributed throughout the world; human infections with some of diphyllobothriids seem to have emerged recently [2 , 3 , 4 , 5] . Causative agents of these diseases are tapeworms of the genera Adenocephalus Nybelin , 1931 , Diphyllobothrium Cobbold , 1858 and Diplogonoporus Lönnberg , 1892 . Out of 16 species of diphyllobothriids reported from man , only the following 4 are common parasites of man , namely Diphyllobothrium latum ( Linnaeus , 1758 ) with circumboreal distribution ( with few cases reported also from Chile ) , D . nihonkaiense Yamane , Kamo , Bylund and Wikgren , 1986 in the northern Pacific region , Adenocephalus pacificus Nybelin , 1931 endemic in the southern Pacific , and Diphyllobothrium dendriticum ( Nitzsch , 1824 ) with arctic distribution [2 , 3 , 4] . These cestodes , commonly called broad or fish tapeworms , produce up to 36 , 000 eggs per day [6 , 7] . Therefore , diagnosis of diphyllobothriosis is based mainly on findings of thick-shelled , unembryonated eggs with an operculum on the narrower end and a knob on the opposite site in stool samples [2] . The morphometry of eggs has been commonly used for species identification of diphyllobothriids in stool samples from naturally infected hosts including man and also experimentally infected unspecific hosts such as golden hamsters ( Mesocricetus auratus ) [8 , 9 , 10 , 11 , 12] . However , most authors used only ranges ( minimum and maximum ) of egg size in descriptions , which overlap among most of the species . Another limitation of previous studies represents a low number of measured eggs and the fact that only three species , D . latum , D . dendriticum and D . nihonkaiense , were studied in more detail . Surface structures on the eggs and some other characters such as shell thickness , colour , diameter of an operculum , presence of an apical knob , or size of embryonic hooks were considered by previous authors , but none of them appeared suitable for species discrimination [13 , 14 , 15 , 16] . In the present study , we analysed so far the largest dataset of diphyllobothriid eggs from naturally and experimentally infected specific hosts , thus significantly improving upon previous studies on egg morphometry . The principal aim was to test whether combination of morphometrical and morphological features by light and/or scanning electron microscopy can make it possible to distinguish reliably the eggs of as many as 8 species of broad tapeworms reported from man .
Gravid tapeworms and/or positive stool samples were obtained from naturally and experimentally infected specific hosts belonging to 19 species , including man; most samples were obtained from museum collections ( see Table 1 ) . They were identified using identification keys based on morphology and mostly also by genotyping ( sequencing the cox1 gene ) [17] . Morphometrical variability was studied in 62 samples of a total of 2 , 082 eggs of 8 species ( Table 1 ) ; these samples were fixed in 70% ethanol , 4% formaldehyde solution or , in few cases , measured alive in the water . Measurements were taken using Olympus BX51 microscope with QuickPHOTO MICRO 2 . 3 program . In every sample , at least 25 intact eggs were measured to get representative dataset [18] . Standard deviation ( SD ) , mean and length to width ratio ( LWR ) were counted using Microsoft Excel software . Summary data for egg measurements ( in micrometres ) pooled across all hosts for 3 morphometrical parameters ( i . e . length , width and LRW; hereinafter referred to as ‘size’ ) are given in Table 2 . A series of univariate comparisons was performed by separate one-way analyses of variance ( One-way ANOVA’s ) for sets of analyses ( A1–A3 –see Table 3 ) in which different numbers of specimens and parasite species were used to assess: ( A1 ) differences in egg sizes ( i . e . 3 morphometrical features: length , width and LRW ) among all 8 diphyllobothriid species; ( A2 ) differences in egg sizes among the 3 commonest human species , namely A . pacificus , D . latum and D . nihonkaiense; and ( A3 ) intraspecific variability in egg sizes of the species from minimally 3 different hosts were available: ( A3a ) A . pacificus , ( A3b ) D . latum , and ( A3c ) D . cordatum . Samples from hamster were not included because it is an atypical host and their eggs are considerably variable compared to specific hosts ( Table 1 ) . Accordingly , 56 samples of a total of 1 , 860 eggs were entered to statistical analyses ( Tables 1 and 3 ) . All data were log-transformed ( log10 ) , but absolute values of egg sizes were used in graphs for better illustration . Post-hoc Tukey HSD tests were performed where appropriate to detect differences in particular morphometrical features among species pairs . All analyses were carried out using Statistica 7 . 0 software package ( StatSoft Inc . , Tulsa , OK , USA ) with significance levels set at 0 . 05 . Additionally , 184 eggs of 44 samples of all 8 studied species were studied using scanning electron microscopy ( SEM ) ( Table 1 ) . Proglottids with eggs were prepared as outlined by Kuchta and Caira [19] . Briefly , after drying the eggs were liberated by dissecting needles from gravid proglottids during mounting on aluminium stubs using a double-sided tape . Samples were examined by JEOL JSM-7401F scanning electron microscope . Surface structures were observed at magnification ×10 , 000 and the density of the pits was re-counted for 100 μm2 . Eggs of diphyllobothriidean cestodes from human samples were obtained from the following museum collections ( in most cases , proglottids found in stool samples were fixed and deposited in the collections; the eggs were obtained as described above ) : Institute of Parasitology , CAS , České Budějovice , Czech Republic ( IPCAS ) , Muséum d’Histoire Naturelle , Geneva , Switzerland ( MHNG ) , Queensland Museum , Australia ( QM ) , and Zoological Museum , University of Oslo , Norway ( ZMUO ) . In addition , samples of eggs from non-human hosts were obtained from museum material deposited in: Natural History Museum , London , UK ( BMNH ) , IPCAS , MHNG , Statens Naturhistoriske Museum , Copenhagen , Denmark ( NHMD ) , Naturhistorisches Museum , Vienna , Austria ( NMW ) , United States National Parasite Collection , Beltsville , Maryland , USA ( USNPC ) , ZMUO ( Table 1 ) .
Morphometrical analysis of all eggs has shown a great size variability of most species studied ( Fig 1 ) . The results of analysis A1 showed significant differences in egg sizes among all 8 species ( P < 10−4 for all 3 morphometrical features tested; Fig 2; Table 3 ) . The post-hoc test showed significant differences between most of the species except D . dendriticum and D . hians , for which egg measurements overlap in length ( P > 0 . 05 ) demonstrating no apparent difference in this parameter ( Fig 2 ) . Diphyllobothrium cordatum has the largest eggs ( mean length 71 ) followed by D . latum ( 68 ) and D . stemmacephalum ( 64 ) , whereas D . cf . cameroni and A . pacificus have the smallest eggs ( 48 and 53 , respectively; Fig s 1 and 2; Tables 1 and 2 ) . Concerning egg width , measurements divided all species into 2 species groups . The first group includes D . cordatum , D . hians and D . stemmacephalum , and second one D . dendriticum and D . cf . cameroni; they did not differ from each other ( P > 0 . 05 ) ( Fig 2; depicted by black and white arrows , respectively ) . The widest eggs were found in D . latum ( mean width 49 ) , followed by D . cordatum ( 47 ) , whereas D . cameroni and D . dendriticum have the narrowest eggs ( 39 and 40 , respectively; Table 2 ) . The post-hoc test revealed that the eggs of D . cf . cameroni and A . pacificus are rounder than those of the 6 remaining species ( mean LWR 1 . 218 and 1 . 272 , respectively; Table 2 ) . This morphometrical parameter significantly differs among all but 2 pairs of species , D . latum and D . stemmacephalum , and D . dendriticum and D . nihonkaiense ( P > 0 . 05 ) , which cannot be distinguished based on this character ( Fig 2; depicted by black and white arrows , respectively ) . One of the 8 species , A . pacificus , can be distinguished from the 7 remaining species by all 3 morphometrical features ( length , width and LWR; Fig 2; Table 3 ) . The remaining species form 5 groups in which a given species overlaps in one particular parameter , but significantly differs in the other two , thus making them distinguishable among each other as well . For example , D . hians and D . dendriticum create the first group by overlapping measurements in length but are different in width and LWR index ( Fig 2; Table 2 ) . Univariate comparisons ( analysis A2 ) of the eggs among the 3 commonest human-infecting species , i . e . A . pacificus , D . latum and D . nihonkaiense , showed significant differences in all 3 parameters tested ( Table 3 ) . However , despite these differences , the range of measurements of D . latum and D . nihonkaiense slightly overlapped in all 3 characters ( Figs 1 and 2; Table 2 ) . The eggs of A . pacificus are smallest ( the lowest mean length and width ) as well as most rounded ( Figs 1 and 2; Table 2 ) . In contrast , D . latum possesses the largest eggs and the eggs of D . nihonkaiense have the highest LWR , i . e . more elongate eggs ( Table 2 ) . The eggs of A . pacificus , D . latum and D . cordatum from various hosts ( analyses A3a–A3c ) differed significantly in all 3 morphometrical parameters ( Table 3 ) . The eggs of A . pacificus from a jackal , Canis mesomelas , were on average longest compared to conspecific eggs from 6 other hosts and differed in width from 2 hosts only ( Fig 3; Table 1 ) . The eggs of A . pacificus from humans and from fur seal , Arctocephalus pusillus , were shortest on average . The widest eggs of A . pacificus were found in sea lion , Neophoca cinerea , and fur seal , Callorhinus ursinus ( Fig 3; Table 1 ) . The most elongate eggs come from Neophoca cinerea and the most rounded from a jackal ( Fig 3 , Table 1 ) . Significant differences were detected in all morphometrical features of the eggs of D . latum from different hosts ( analysis A3b; Table 3 ) . Eggs from humans and polar bear were longest and widest compared to the eggs from dog and wolf ( Fig 3; Table 1 ) . Eggs from human stool samples were also most elongate ( Table 1 ) . The eggs of D . cordatum from 3 different hosts also differed significantly from each other in all parameters ( P < 0 . 001; Fig 3; Table 3 ) . The eggs from dogs were smaller compared to those from seals ( Fig 3 ) whereas most elongate eggs were from Erignathus barbatus ( Table 1 ) . Eggs of all marine species ( Adenocephalus pacificus , D . cf . cameroni , D . cordatum , D . hians and D . stemmacephalum ) differ conspicuously from those of freshwater/anadromous species ( D . dendriticum , D . latum and D . nihonkainese ) in the presence of numerous deep pits on their surface observed using SEM or light microscope with a high magnification and strong pressing coverslip with egg samples ( Figs 4 and 5 ) [14 , 20] . Density of pits of marine species differed between species , from 34 pits per 100 μm2 in A . pacificus to as many as 207 in D . cordatum ( Fig 4; Table 1 ) . This characteristic may help as an additional feature to distinguish species groups with similar egg sizes , such as D . hians from those of D . stemmacephalum ( Figs 2 and 4; Tables 1 and 2 ) . The eggs of the remaining freshwater/anadromous species are indistinguishable from each other based on their surface , because it is smooth or occasionally covered with a few shallow hollows or wrinkles ( Fig 4; Table 1 ) . To facilitate species diagnosis of eggs in stool ( clinical ) samples , a simplified key for identification of the eggs of 8 human-infecting species of diphyllobothriid cestodes based on their length and width is provided . The mean value represent means between the species from each compared groups with closest values , i . e . the smallest value in the larger group and largest values in the smaller group and species of the genus Diplogonoporus are not considered ( see Discussion ) . The present key does not include data on pits observed on the eggs from jackal as an unspecific host ( see Table 1 ) .
Human diphyllobothriosis is not a life-threating disease and symptoms are usually mild despite the large size of tapeworms ( up to 20 m ) . However , the number of human case does not decline , reaching about 20 million cases globally [21] . In contrast , several species seem to ( re- ) emerge or neglected even in the most developed countries [2 , 3 , 22] . For clinicians , species-specific identification of eggs in stool samples is not necessary because clinical symptoms are similar in all human-infecting species [2] . However , proper species identification is crucial from the epidemiological point of view , i . e . to detect sources of human infection , actual distribution of potential human-infecting species and way of transmission of plerocercoids to humans [2] . The present study represents the most complex statistical analysis based on the so far largest dataset for diphyllobothriid eggs . The main novelty of this study is combination of morphometrical and ultrastructural characteristics of diphyllobothriid eggs which enabled us to distinguish individual human-infecting species from each other . Species-specific identification of eggs is thus possible despite a rather high degree of their morphometrical and morphological variability in most of the studied species observed in the present and previous studies ( see below ) . Previously , the most extensive study of diphyllobothriid eggs was carried out by Hilliard [14] who studied 8 species from naturally and experimentally infected hosts in Alaska and by Andersen and Halvorsen [10] who compared the 3 commonest freshwater species in Europe , D . ditremum , D . dendriticum and D . latum , from a variety of definitive hosts . Despite some differences , especially in the size of the eggs of D . cordatum from bearded seal , E . barbatus [14] , the present data correspond to those provided by previous authors [10 , 14 , 23 , 24] . Additional study based on 7 species from human clinical samples studied by Maejima [25] focused mainly on eggs of D . latum , D . nihonkaiense , A . pacificus and D . stemmacephalum ( reported as D . yonagoense ) . These data correspond more or less with those of the present study , even though values reported for the 2 former species were somewhat lower than herein . The eggs of D . latum have been studied most intensively , especially those from man; a high degree of size variability ( 55–81 × 40–59 ) has been detected [10 , 26; present study] . However , some of these records may in fact have included other species such as D . nihonkaiense in Far East Asia and D . dendriticum in temperate zones [2 , 12 , 27] . The detailed study comparing the 3 commonest freshwater species in Europe , D . ditremum , D . dendriticum and D . latum , from different definitive hosts was carried out by Andersen and Halvorsen [10] . They observed great variation in mean egg size among different worms belonging to the same species as well as among different species . It was concluded that size of the eggs also depends on the size and type of definitive hosts [10 , 11 , 28] . The intraspecific variability based on host species was compared in 3 species in the present study . Significant differences were found in all 3 species , but their ranges overlapped in almost all host groups ( Fig 3 ) . In general , there is tendency of tapeworms from larger specific hosts to have larger eggs , but that scenario is not confirmed by all studies . For example , the present study shows that eggs of A . pacificus from man ( atypical host ) are significantly smaller than those from pinnipeds ( specific hosts ) but , surprisingly , the longest eggs are those from jackal , which is an atypical host ( Fig 3; Table 1 ) [17] . Moreover , some previous studies did not find significant differences between the size of eggs of A . pacificus from man and those from fur seals , even though the eggs from man were slightly smaller and the surface was covered with more pits [11 , 28] . Eggs from atypical experimental hosts such as golden hamster may be larger than those from typical hosts as observed in D . dendriticum or D . latum [10; Table 1] . The present study of egg surface confirmed differences between freshwater and marine species reported by previous authors [14 , 29 , 30 , 31] . The surface of the eggs of all marine species studied is covered with numerous deep pits , whereas freshwater species have a smooth surface or only a limited number of wrinkles or shallow pits ( Fig 4 ) . It is important to point out that only clean eggs should be used for SEM observations to avoid misinterpretation of artifacts as natural surface structures . The egg surface can be observed also by light microscope if eggs are strongly compressed under coverslip and observed ( Fig 5 ) [14 , 20] . Future laboratory experiments with eggs of marine and freshwater species are necessary to provide an explanation this difference in surface structure . The eggs of diphyllobothriid cestodes are heavy and cannot float in the water; instead they sink after their shedding , but the presence/absence of numerous and deep pits on their surface seem to be related to different physical properties of fresh and sea water and certainly plays some , yet unknown role in facilitating parasite transmission [14 , 29] . Combination of morphometry and surface ultrastructure of the eggs appeared to be helpful in identification of 8 human-infecting diphyllobothriideans , including D . latum , D . dendriticum and D . nihonkainse , which are very similar in morphology and morphometry of their adults and eggs [2 , 12] . However , this is possible only if a sufficient number of the eggs ( at minimum 25 ) is measured to avoid statistical error due to a high variability of eggs and to enable reliable species-specific identification [18] . The eggs of the common species A . pacificus can be identified most easily because they are smallest and their surface is covered , similarly as in other marine species , with numerous pits ( Figs 2 and 4; Table 1 ) . However , we are well aware that such a detailed morphometrical and morphological analysis that include preferentially scanning electron microscopy is not possible during routine diagnosis of the stool samples in clinical laboratories . Therefore , positive clinical samples , especially in non-endemic areas , should be always fixed first with ethanol and later identified by molecular methods , mainly by sequencing of the mitochondrial cox1 gene or by available multiplex PCR [2 , 31] . In clinical samples , eggs of other helminths of similar shape and size can be found . The most similar are those of another diphyllobothriideans normally infecting whales , Diplogonoporus balaenopterae Lönnberg , 1892 , which causes human diplogonoporosis [2] . The eggs of this cestode from human samples are 57–80 long and 34–49 wide; their surface is covered with numerous pits ( 150–250 pits per 100 μm2 ) [14 , 20 , 29 , 30] . They can thus be confused with those of Diphyllobothrium cordatum and D . hians ( see Table 2 ) . However , the number of human cases of diplogonoporosis is quite low , around 200 worldwide [2] . To conclude , the present study provides evidence that combination of several characteristics assessed by statistical methods represents a useful tool to differentiate otherwise indistinguishable eggs of human-infecting broad fish tapeworms . Even though a detailed morphometrical and morphological ( ultrastructural ) characterisation of the diphyllobothriid eggs is not trivial , but relatively fast and cheap and could be used for routine diagnostics . Accurate identification of the species causing diphylloobothriosis is essential for understanding of the epidemiology and transmission of this neglected fish-borne human disease , which seems to ( re- ) emerged due to changing eating habit even in the most developed countries . | More than 2 , 000 eggs of 8 species of diphyllobothriid cestodes infecting humans were compared . Combination of morphometrical and ultrastructural ( surface morphology ) data made it possible to distinguish all species . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"cestodes",
"light",
"microscopy",
"parasitic",
"diseases",
"animals",
"morphometry",
"microscopy",
"fresh",
"water",
"research",
"facilities",
"research",
"and",
"analysis",
"methods",
"hydrology",
"imaging",
"techniques",
"flatworms",
"scanning",
"electron",
"microscopy",
"pathogenesis",
"museum",
"collections",
"host-pathogen",
"interactions",
"earth",
"sciences",
"electron",
"microscopy",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2016 | Eggs as a Suitable Tool for Species Diagnosis of Causative Agents of Human Diphyllobothriosis (Cestoda) |
Mutations in the human Zip4 gene cause acrodermatitis enteropathica , a rare , pseudo-dominant , lethal genetic disorder . We created a tamoxifen-inducible , enterocyte-specific knockout of this gene in mice which mimics this human disorder . We found that the enterocyte Zip4 gene in mice is essential throughout life , and loss-of-function of this gene rapidly leads to wasting and death unless mice are nursed or provided excess dietary zinc . An initial effect of the knockout was the reprogramming of Paneth cells , which contribute to the intestinal stem cell niche in the crypts . Labile zinc in Paneth cells was lost , followed by diminished Sox9 ( sex determining region Y-box 9 ) and lysozyme expression , and accumulation of mucin , which is normally found in goblet cells . This was accompanied by dysplasia of the intestinal crypts and significantly diminished small intestine cell division , and attenuated mTOR1 activity in villus enterocytes , indicative of increased catabolic metabolism , and diminished protein synthesis . This was followed by disorganization of the absorptive epithelium . Elemental analyses of small intestine , liver , and pancreas from Zip4-intestine knockout mice revealed that total zinc was dramatically and rapidly decreased in these organs whereas iron , manganese , and copper slowly accumulated to high levels in the liver as the disease progressed . These studies strongly suggest that wasting and lethality in acrodermatitis enteropathica patients reflects the loss-of-function of the intestine zinc transporter ZIP4 , which leads to abnormal Paneth cell gene expression , disruption of the intestinal stem cell niche , and diminished function of the intestinal mucosa . These changes , in turn , cause a switch from anabolic to catabolic metabolism and altered homeostasis of several essential metals , which , if untreated by excess dietary zinc , leads to dramatic weight loss and death .
The rare pseudo-dominant genetic disease acrodermatitis enteropathica ( AE ) is thought to be caused by the inefficient absorption of dietary zinc [1] . AE occurs at a frequency 1 in 500 , 000 and symptoms in humans usually develop soon after birth in bottle fed infants or after weaning in breast fed infants . The triad of alopecia , eczematous dermatitis and diarrhea are classic symptoms of AE in humans and patients experience growth retardation and a myriad of symptoms of severe zinc deficiency which eventually lead to death unless treated by exogenous zinc [2] . In 2002 , the AE gene was mapped to human chromosomal region 8q24 . 3 and the Zip4 ( Slc39a4 ) locus [3] , [4] . Currently , over 32 mutations or variants of ZIP4 have been reported [5] . Missense and nonsense mutations as well as deletions or rearrangements of the gene have all been reported , and hypomorphic as well as complete loss-of-function alleles have been identified . Recent studies have shed light on the mechanisms of Zip4 regulation and function [6] . Mouse Zip4 is most actively expressed in tissues involved in the absorption of dietary or maternal zinc , but also shows high level expression in other cell-types ( e . g . pancreatic islet cells , brain capillaries ) , and low level expression in other tissues ( e . g . liver , kidney ) and some cultured cells . Zip4 expression is regulated by cell-specific transcription as well as by multiple posttranscriptional mechanisms in response to zinc availability . ZIP4 protein is at the apical surface of enterocytes and endoderm cells when zinc is deficient , due to increased mRNA and protein stability . During zinc deficiency ZIP4 undergoes processing by removal of the extracellular amino-terminus . In contrast , in the presence of normal levels of zinc Zip4 mRNA is unstable and the protein is internalized and rapidly degraded . ZIP4 function is critical during periods of rapid growth when zinc requirements are high but this zinc transporter also has important functions when zinc is replete . Zip4 is aberrantly expressed in many cancers [7] , [8] . Knockdown of ZIP4 can slow cell cycle and cell migration in mouse Hepa cells and ZIP4 functions to reduce apoptosis and enhance cell cycle in hepatomas and to enhance pancreatic tumor growth in nude mice [7] , [8] . Many recent studies have demonstrated that zinc can modulate signal transduction cascades [9] . The essential function of ZIP4 in zinc homeostasis is confirmed in Zip4-intestine knockout mice . Homozygous Zip4-intestine knockout mouse embryos die soon after implantation because this gene is also actively expressed in the visceral endoderm which surrounds the developing mouse embryo and which serves a nutrient uptake function before development of the placenta . The visceral yolk sac is a more vestigial organ in humans and apparently contributes little to providing nutrients for the embryo , explaining why loss of ZIP4 in humans is apparently not embryonic-lethal . Interestingly , heterozygous Zip4-intestine knockout mouse embryos are significantly underrepresented in the population at parturition . These heterozygous embryos display an array of developmental defects including exencephalia , anophthalmia , and severe growth retardation . About 22% of the Zip4-heterozygous mice that survive to weaning age are severely growth retarded and display anopia , anopthalmia , hydrocephalus and heart defects , among other abnormalities [10] . Mice heterozygous for Zip4 knockout are hypersensitive to zinc deficiency . Thus , haploinsufficiency of Zip4 may contribute to growth retardation in humans , an effect that is probably exacerbated by zinc deficiency and/or by modifier genes . Mutations in genes essential for posttranscriptional regulation of ZIP4 may also cause AE although this has not been demonstrated . The finding of Zip4 haploinsufficiency defines AE as a pseudo-autosomal dominant trait . Given that a global knockout of Zip4 is embryonic lethal in mice , unlike it is in humans , we sought to develop a better mouse model of AE . To that end we created mice with floxed Zip4 genes and bred them with mice that express an ErtCre fusion protein driven by the villin promoter [11] specifically in intestine enterocytes . Using this tamoxifen-inducible Zip4-enterocyte knockout model we provide evidence that intestine expression of Zip4 is essential for the growth and viability unless mice are supplied with excess zinc and that an absence of ZIP4 in the intestine appears to closely mimic the AE phenotype in humans . Moreover we provide evidence that an absence of ZIP4 only in the intestine leads to a rapid switch from anabolic to catabolic metabolism in the animal , to tissue-specific dysregulation of other essential metals and alterations in gene expression . These phenotypes appear to reflect compromised Paneth cell functions which lead to disruption of the intestinal stem cell niche ultimately resulting in loss of intestinal integrity and diminished nutrient uptake .
To enable tissue-specific deletion of the mouse Zip4 gene a targeting construct was created which contained a LoxP site flanked by an XhoI restriction site in intron 5 , and a LoxP site just downstream of the last exon ( exon 12 ) , followed by an mc1-Neomycin cassette ( Figure 1 ) . This construct was targeted in E14 embryonic stem ( ES ) cells and cells with proper integration of the floxed Zip4 gene were then identified by long-range PCR using primers outside of the engineered targeting construct coupled with overlapping internal primers ( Figure 1A ) . The 5′ integration screen ( Figure 1C ) amplified a 7 . 35 kb product from the wild-type and the floxed alleles and cleavage of the floxed allele with XhoI yielded the predicted 5 . 2 and 2 . 1 kb restriction fragments indicative of proper insertion of the floxed allele into one of the endogenous Zip4 alleles . This was confirmed by the 3′ integration screen which yielded a 3 . 4 kb PCR product from the floxed allele and a 2 . 1 kb PCR product from the wild-type allele . Targeted E14 ES cells were used to create chimeric mice by blastocyst injection and agouti offspring from the chimeric mice were genotyped using primers which flank the LoxP-XhoI insertion site in intron 5 ( Figure 1A ) . Genotyping PCR yielded a 227 bp product from the floxed allele , which could be cleaved by XhoI , and a 187 bp product from the wild-type allele ( Figure 1D ) . To enable tissue-specific deletion of the floxed Zip4 gene specifically in the intestinal epithelium , Zip4FX/FX mice were crossed with mice that express an estrogen receptor-Cre recombinase fusion protein under control of the villin promoter [11] . Vil-Cre-ERT2 is expressed only in the intestinal epithelium and visceral endoderm . The Cre-ERT2 fusion protein resides in the cytoplasm and is driven to the nucleus by tamoxifen which allows for temporal control of the Zip4 gene deletion [11] . Mice heterozygous for vil-Cre-ERT2 and homozygous Zip4FX/FX were crossed with Zip4FX/FX mice to yield 50% offspring with Zip4FX/FX: vil-Cre-ERT2 alleles and 50% with Zip4FX/FX alleles . The latter provided age and genetically matched controls for all of our experiments and are labeled as control ( Con ) mice in the figures . Tamoxifen injections resulted in recombination of approximately 50% of the floxed Zip4 genes in the small intestine of recently weaned Zip4FX/FX: vil-Cre-ERT2 mice ( Figure 1D: intest KO ) . This is as expected since the epithelial cells represent about half of the cells in the small intestine . The extent of loss of Zip4 mRNA however provides a suitable assay for the extent of recombination of the floxed Zip4 gene in the epithelium since the expression of this gene in the intestine is restricted to the epithelium . Three injections of tamoxifen resulted in a dramatic , if not complete loss of intestine Zip4 mRNA ( Figure 1E: intest KO ) which suggests that the vast majority , if not all , of the floxed enterocyte Zip4 genes had undergone recombination . ZIP4 protein in the intestine of mice fed a zinc-adequate diet is present at only low levels but this mRNA and protein are stabilized when zinc is deficient . ZIP4 protein and mRNA are rapidly turned-over when dietary zinc is restored . The loss of Zip4 mRNA is expected to be followed by the turn-over of ZIP4 protein within a few hours [12] . Symptoms of AE in humans often appear soon after breast feeding is stopped . Human breast milk provides a rich source of zinc that is easily absorbed by the gut [13] . The effect of knocking out intestine Zip4 in neonatal mice ( 5 days post-partum ) was therefore examined . Newborn mice were injected with tamoxifen and allowed to suckle until weaning on day 21 post-partum ( Figure 2A ) . At weaning there was no difference in the average body weight of the control ( 7 . 93 g: n = 4 ) and knockout mice ( 7 . 96 g: n = 5 ) and the intestine knockout mice appeared normal . However , the Zip4-intestine knockout mice rapidly lost weight after weaning ( Figure 2A and Figure S1A ) and perished within a week unless provided with excess zinc in the drinking water ( Figure 2B ) . Removal of the excess dietary zinc led to precipitous weight loss and lethality . Northern blot hybridization confirmed that the Zip4 gene was effectively knocked out in these neonatal mice ( Figure 2B , inset ) . Not all of the recently weaned Zip4-intestine knockout mice thrived after providing excess zinc at weaning . Those mice could be rescued from lethality for three weeks post-weaning but failed to gain much body weight during that period ( Figure 2B ) . The function of intestine Zip4 in rapidly growing , recently weaned mice was then examined . Recently weaned mice ( 5 days post-weaning ) were injected three consecutive days with tamoxifen and their body weights were measured beginning 1 day after the first injection ( Figure 2C and Figure S1B ) . These mice continue to grow for two to three days after the first injection and then begin to lose weight precipitously . Within a week they lose over 20% of their body weight although they continue to eat a normal amount of food and by d8 they can weigh only 60% as much as their littermate controls ( Figure 2D ) . On day 6 the tibia and quadriceps muscle from the Zip4-intestine knockout mice weigh 43 to 50% less than controls while body weight has decreased 23 to 30% relative to littermate controls ( n = 4–5; p<0 . 007 ) . Much of the weight loss involves loss of muscle and bone mass as has been reported to occur during dietary zinc deficiency [14] . Thus , knocking out the intestine Zip4 gene causes a switch from anabolic to catabolic metabolism in these mice . During the course of these studies we examined many mice and all of the Zip4-intestine knockout mice died within two to three weeks . The penetrance of this phenotype is 100% in both male and female mice . Northern blot hybridization to small intestine RNA demonstrated a loss of Zip4 mRNA and a dramatic reduction of Zip5 mRNA ( Figure 2C and 2E , insets ) . ZIP5 is a basolateral zinc transporter expressed predominately in intestinal crypt cells [15] . This mRNA remains unchanged in abundance during severe zinc deficiency in mice [15] but its translation is repressed [12] by a mechanism involving a stem-loop structure in its mRNA [16] . Attenuation of Zip5 mRNA , whose abundance is not responsive to Zn itself , could suggest injury to the small intestine crypts , as is shown below . Excess dietary zinc can ameliorate many of the symptoms of AE in humans . Therefore , control and Zip4-intestine knockout mice were provided drinking water containing 250 ppm zinc sulfate beginning on the day of the first tamoxifen injection ( Figure 2E ) . A representative animal is shown but 12 mice were analyzed with similar results . Excess zinc prevented the rapid weight loss in the Zip4-intestine knockout mice and allowed them to thrive for 4 weeks in this experiment , but withdrawal of the excess zinc led to a rapid loss of weight and lethality within two weeks . The Zip4 gene remained knocked out for 4 weeks in these mice ( Figure 2E , inset ) which indicates that it was also knocked out in the intestinal stem cell population , as reported previously using the vil-CreERT2 system [11] . In other experiments it was noted that providing excess zinc could not effectively ameliorate the lethal effects of the Zip4 knockout if provided later than day 6 ( data not shown ) . The efficacy of one , two or three consecutive injections of tamoxifen on Cre-ERT2 deletion of the Zip4 gene was examined using recently weaned mice . As described above , three consecutive injections resulted in effective deletion of the Zip4 gene even in the intestinal stem cells . In contrast , a single injection of tamoxifen resulted in partial deletion of the Zip4 gene ( Figure 2F , also see Figure S1C ) . This did not result in lethality; instead those mice grew normally for about a week and then began to slowly lose weight . Thereafter , about half of the mice examined recovered vigorous growth and grew to adult weight ( two are shown in Figure 2F ) while the other half of these mice continued to slowly lose weight during that time ( Figure S1C ) . The resumption of vigorous growth coincided with active expression of Zip4 mRNA which had returned to normal levels by three weeks ( Figure 2F , inset ) . These data show that partial loss-of-function of ZIP4 can exert detrimental effects of growth consistent with the findings of potentially hypomorphic alleles of Zip4 in AE [17] and that heterozygosity of Zip4 renders mice hypersensitive to zinc deficiency [10] . Many patients with AE require lifelong zinc supplementation . Thus , ZIP4 may function not only during periods of rapid growth but also in full grown adults . Adult mice ( 7 weeks old ) were injected with tamoxifen and then monitored for changes in body weight ( Figure 2G ) . The knockout mice began to lose weight within days of the first injection , lost weight precipitously after 1 week and then succumbed soon thereafter . Thus , the loss of intestine ZIP4 led to a switch from anabolic to catabolic metabolism in adult mice . ZIP4 is a zinc-specific transporter [18] . Although a recent study suggests that it may also be able to transport small amounts of free copper the physiological relevance of that activity remains to be determined [19] . Herein the effect of knocking out intestine Zip4 in recently weaned mice , on the homeostasis of several essential metals was examined using ICP-MS . Within 4 days of initiating the knockout , zinc levels in the small intestine , pancreas and liver were significantly reduced ( Figure 3A ) consistent with a primary function of ZIP4 in the acquisition of dietary zinc . This loss of zinc correlated temporally with the beginning of precipitous body weight loss after the knockout . Although , by day 4 a small but statistically significant loss of copper ( Figure 3B ) also occurred in these organs , the largest copper loss was in the small intestine which also lost ∼50% of its iron ( Figure 3C ) . Thus , the loss of ZIP4 in the intestine led to large reductions in small intestine zinc , iron and copper content suggesting possible injury to the intestine , as was found to be the case . An increase in manganese was noted in the liver ( Figure 3D ) , but no other significant changes in other elements in these tissues were found on day 4 . Remarkably , by day 8 after initiation of the knockout , iron and manganese were twice as high in the liver of Zip4-intestine knockout mice versus control and hepatic copper had also increased significantly ( Figure 3E ) . In contrast , zinc levels were normalized in the liver which suggests that zinc had been redistributed as body weight declined . The accumulation of iron in the liver during zinc deficiency has been reported previously [20] , [21] . Taken together , these results show that intestine ZIP4 is essential for the homeostasis of several essential metals . Being able to temporally control the intestine knockout of Zip4 enables us for the first time to monitor the rapid effects of loss-of-function of a zinc transporter in an animal . This model of AE compresses the time frame from months in humans to a few days in mice . This allows for examining primary versus secondary effects of losing Zip4 function . To that end Northern blot hybridization and quantitative PCR were used to monitor changes in Zip4 , Zip5 and metallothionein- I ( MT-I ) mRNAs , zinc homeostatic genes , in the small intestine ( Figure 4 ) . In addition changes in IGF-1 mRNA , a potent mitogen in the intestine [22] , were monitored . A single injection of tamoxifen resulted in a large loss of native Zip4 mRNA in the small intestine 24 hr later ( Figure 4 ) . By one day after the second injection ( day 2 ) , little or no native Zip4 mRNA was detectable . Zip5 mRNA abundance dropped precipitously by day 3 while MT-I mRNA was reduced in abundance within a day of the first tamoxifen injection but increased again by d4 and remained high ( Figure 4 ) . Thus , deletion of the intestine Zip4 gene occurs rapidly in this model and leads to significant changes in the abundance of several mRNAs involved in zinc homeostasis . The loss of Zip5 expression suggests that injury to the intestine may occur rapidly after deletion of the Zip4 gene , as Zip5 mRNA abundance is not zinc-responsive . Deletion of the intestine Zip4 gene led to a large increase in the abundance of IGF-I mRNA . Several IGF-I transcripts are present in the intestine . Quantitative PCR of the region encoding IGF-I in all of the reported transcripts confirmed this Northern blot finding and established that this mRNA is induced up to 7-fold within a few days ( Figure 4 ) . There were no changes in the abundance of IGF-I transcripts in control littermates ( Figure 4 , Con ) . The Con blot was over-exposed relative to that of the knockout mice . Histological examination of the small intestine after Zip4 knockout revealed progressive and profound disorganization of villus ( Figure 5 ) and crypt architecture ( Figure 6 ) and significantly diminished cell division within days of the knockout ( Figure 7 ) . Villus histology was apparently normal on day 2 after initiation of the knockout but became progressively disorganized over the next 4 days ( Figure 5A ) . By day 6 the epithelial cells had lost much of their columnar morphology with basal nuclei and had become cuboidal with centric nuclei ( Figure 5B ) . The lamina propria appeared disorganized , occupied substantially more space in the villi and often contained red blood cells and cells with pyknotic nuclei . Crypts were examined by localizing the transcription factor Sox9 in sections of small intestine . Sox9 ( sex determining region Y- box9 ) marks the stem/progenitor cell population in the crypts [23] as well as the villus enteroendocrine cells [24] and it regulates cell proliferation and Paneth cell differentiation [25] . In control mice and Zip4-intestine knockout mice before tamoxifen injections and on day 2 after initiation of the knockout , Sox9 was localized to the crypts ( Figure 6A ) and to dispersed enteroendocrine cells in the villi ( not shown ) , as reported previously [24] . However , after day 2 the architecture of the crypts progressively became disorganized and elongated , and Sox9 positive cells were increasingly detected extending up into the villus epithelium with Sox9 negative cells interspersed among the positive cells ( Figure 6A ) . By days 4 and 6 , Sox9 negative cells were found at the base of many crypts ( Figure 6A and 6C ) . The location of these cells at the base of the crypts suggested that these were Paneth cells; however , these cells contained large cytoplasmic vacuoles not found in normal Paneth cells . Paneth cells contain large amounts of labile zinc which can be detected by staining with the zinc-specific fluorophore ZP1 [26] . ZP1 staining was readily detectable in the intestinal crypts before initiation of the Zip4 knockout ( Figure 6B ) . When viewed under high power ZP1 staining was restricted to one or two cells at the base of the crypts ( Figure 6B; 400X panel ) whereas under low power the fluorescence signal was dispersed throughout the crypt . ZP1 staining was lost by day 2 after initiation of the knockout and remained undetectable in the Paneth cells on day 4 . Thus , zinc is rapidly depleted in Paneth cells when ZIP4 function is lost . The antibacterial enzyme lysozyme is abundantly produced by Paneth cells [27] . IHC localization of lysozyme in intestine sections from control and Zip4-intestine knockout mice confirmed that Paneth cells accumulate large vacuoles , and have reduced Sox9 and zinc when Zip4 is knocked out ( Figure 6C ) . Paneth cells , enteroendocrine and goblet cells are thought to arise from a common precursor [28] . Paneth cells constitute the niche for Lgr5-positive columnar stem cells in the crypt base [29] , [30] . The Lgr5-positive columnar cells can give rise to all epithelial lineages [29] . PAS staining of mucins revealed normal goblet cells interspersed in the villus epithelium of uninjected and knockout mice ( Figure S2 ) . PAS staining of the small intestine from Zip4-intestine knockout mice demonstrated positive staining of the large vacuoles in Paneth cells by day 4 ( Figure 6C ) . This staining was resistant to α-amylase and was therefore not due to glycogen ( data not shown ) . Enteroendocrine ( Sox9+ ) cells and goblet ( PAS+ ) cells are long-lived in the villus epithelium and appeared to be normal in numbers after knocking out intestine Zip4 , at least within the time period examined ( Figure S2 ) . No change in Paneth cells were seen in control littermates ( Con ) that had been similarly injected with tamoxifen and processed in parallel ( data not shown ) . These results indicate that loss-of-function of intestine ZIP4 leads to a rapid loss of Paneth cell zinc which is followed by changes in Paneth cell morphology and phenotypic reprogramming . The relative amount of cell division in the small intestine was determined since Sox9 plays a role in intestinal stem cell proliferation and the crypts are critical for renewal of the epithelium . Cell division was monitored by IHC for phosphorylated ( Ser 10 ) histone H3 and quantified by counting positively stained cells in multiple fields of view ( 5 or more ) in sections from several ( 4 or 5 mice ) control and Zip4-intestine knockout mice ( Figure 7 ) . On day 2 after initiation of the knockout , cell division was measurably repressed ( 22% ) in the small intestine and by day 4 , cell division was reduced by 47% and remained repressed on day 6 . No reduction in cell division after tamoxifen injections was noted in control littermates ( data not shown ) . These results reveal that intestine ZIP4 functions to maintain the structural and functional integrity of the small intestine . Given the repressive effects of loss-of-function of ZIP4 on small intestine cell division and structural integrity leading to a switch from anabolic to catabolic metabolism in mice , we examined the effects of this knockout on the activity of mammalian target of rapamycin ( mTOR1 ) in the intestine . mTOR1 plays an important role in growth regulation and intestinal cell migration by integrating signals from nutrients , energy status , growth factors and stresses thereby controlling protein translation [31]–[33] . Ribosomal protein S6 kinase is activated by mTOR1 leading to phosphorylation of ribosomal protein S6 and enhanced translation [31] . Therefore , the status of phosphorylated ribosomal protein S6 was monitored by IHC in the small intestine of control and Zip4-intestine knockout mice ( Figure 8 ) . Phosphorylated ribosomal protein S6 was readily detectable in the enterocytes of the villus tips in control mice . Little or no staining was detected in crypts ( Figure 8A ) . However , by d2 after initiating the knockout staining in the crypts became apparent while that in the villus tips was diminished ( Figure 8B ) , and by d4 crypt staining was strong and villus staining was weak or absent . On day 6 crypt staining remained strong and cells interspersed in the villus epithelium stained positive for phosphorylated ribosomal S6 protein . Closer examination of the stained crypts revealed an absence of staining in the Paneth cells ( Figure 8B ) . These results indicate that loss of ZIP4 function in enterocytes leads to a rapid and prolonged diminution in mTOR1 activity and impaired protein synthesis in those cells . Remarkably this is accompanied by enhanced mTOR1 activity in the stem cell niche , but not in Paneth cells . Unlike the intestine , the liver in Zip4 intestine knockout mice apparently does not undergo obvious morphological changes and the liver to body weight ratio remains relatively constant during the first week or so after initiating the knockout ( data not shown ) . In contrast there is a rapid loss of bone and muscle mass and body weight in these mice . The liver plays a critical role in metabolism; therefore , the status of AMP-activated protein kinase ( AMPK ) and mTOR1 activity in the liver were examined by Western blotting ( Figure 9 ) . The activity of AMPK is allosterically increased by AMP binding which leads to phosphorylation and activation of AMPK . When ATP generation is attenuated , AMPK becomes activated and phosphorylates many substrates leading to increased activation of catabolic pathways and reduced activation of anabolic pathways [32] . As mentioned above , mTOR1 activity enhances anabolic metabolism by supporting translation and transcription . Western blot analysis of liver proteins revealed that phosphorylated ( T172 ) -AMPK increased soon after the knockout of intestine Zip4 reaching a peak on day 3 before rapidly declining ( Figure 9 ) . This decline was accompanied by increased and prolonged phosphorylation ( S2448 ) of mTOR1 . In contrast , little change was noted in the phosphorylation of AKT ( S473 ) or active β-catenin until day 8 . Taken together , these results show that the loss-of-function of intestine ZIP4 rapidly causes a switch from anabolic to catabolic metabolism in the liver reflecting an increased AMP/ATP ratio . However , this effect was transient and mTOR1 activity was restored when the mice began wasting muscle and bone which would provide amino acids and other energy sources for liver metabolism . Thus , unlike the small intestine ( as well as bone and muscle ) the liver is only transiently affected by the loss-of-function of intestine ZIP4 .
Herein we describe the creation of a mouse model of the rare pseudo-dominant human genetic disorder of zinc metabolism AE . Being able to control the timing and extent of the knockout of enterocyte Zip4 in mice , and the rapid progression of the disease thereafter , allowed us for the first time to effectively address underlying mechanisms of AE . A central remaining question in our understanding of AE has been to define the role of intestinal Zip4 in the etiology of this disease . Mouse Zip4 expression can be detected in several cell-types [18] , [34] . Human Zip4 expression is also detected in several organs ( NCBI Unigene EST profile ) and fibroblasts from AE patients show diminished zinc uptake and zinc content leading to the suggestion that loss-of-function of ZIP4 in organs other than the intestine could contribute to the pathology of AE [35] . Our data demonstrate that the most profound phenotypes of AE , severe growth retardation and morbidity , reflect a loss-of-function of ZIP4 only in intestinal enterocytes , that was predicted from studies of patients with AE nearly 30 years ago [2] . A significant difference between this mouse model and human AE is the lack of severe dermatitis and alopecia in mice . This likely reflects the fact that Zip4 is actively expressed in human skin ( NCBI Unigene EST profile ) whereas it is not actively expressed in mouse skin [18] . In mice , growth retardation and/or weight loss occur rapidly after enterocyte Zip4 is knocked out and the disease progresses to morbidity within a few days accompanied by a switch from anabolic to catabolic metabolism , unless mice were nursing or given access to high zinc concentrations in their water . Similarly , humans with AE respond favorably to increased dietary intake of zinc . How zinc is absorbed from the gut in the absence of ZIP4 is not known . However , uptake of zinc by L-type calcium channels has been reported [36] and several other Zip and ZnT proteins are expressed in the intestine and could contribute to zinc uptake when zinc levels are very high [37]–[39] . Our studies revealed that one of the most rapid effects of loss-of-function of intestine ZIP4 is the loss of labile zinc from Paneth cells . This may therefore be an initial event in the progression of AE . Studies of patients with AE , also reported nearly 35 years ago , revealed changes in Paneth cells including the accumulation of abnormal inclusion bodies and giant granules , which were reversed when patients were given oral zinc [40]–[42] . Zinc deficiency or prolonged fasting has been shown to cause the loss of Paneth cell zinc , as measured by reactivity with the zinc chelator dithizone [43] and dithizone injection causes selective killing of Paneth cells [44] . However , zinc deficiency alone in rats does not lead to abnormal Paneth cell morphology [45] which suggests that factors in addition to zinc loss may contribute to the pathology of Paneth cells in AE patients . The loss-of-function of intestine ZIP4 rapidly alters the phenotype of Paneth cells which are long-lived , suggesting that preexisting Paneth cells undergo reprogramming . The rapid loss of zinc in these cells is followed by reduction of Sox9 and the accumulation of mucin positive Paneth cells . Sox9 regulates cell proliferation in the small intestine epithelium as well as Paneth cell differentiation and repression of mucin ( Muc2 ) expression [23] , [46] . Sox9 is regulated by the Wnt pathway [23] which suggests that ZIP4 may play a role in regulating signal transduction in Paneth cells via the Wnt-APC pathway . A role of zinc in controlling glycogen-synthase kinase-3β ( GSK-3β ) activity and , by extension the Wnt-APC pathway , has been reported [47]–[49] . Increased zinc leads to increased phosphorylation of GSK-3β , the accumulation of β-catenin and increased signal transduction . Therefore , our data are consistent with the concept that the loss of zinc in Paneth cells leads to reduced phosphorylation of GSK-3β , increased turn-over of β-catenin , reduced Sox9 expression , and reprogramming of crypt Paneth cells . Loss-of-function of ZIP4 also resulted in a reduction in cell division in the crypts which may be a consequence of the altered Paneth cell phenotype and crypt dysplasia . A recent report revealed that Paneth cells contribute to the intestinal stem cell niches by expressing several growth factors , most importantly Wnt , which support those stem cells [30] . Perturbed function of Paneth cells would negatively affect Lgr5 columnar stem cells which are interspersed between Paneth cells in the crypts [29] , [30] . These stem cells are able to give rise to all the epithelial lineages in vivo , can generate crypt like structures in vitro , and there is a strong correlation between the number of Paneth cells and the number of Lgr5 stem cells in the intestine in vivo [29] , [30] . Therefore , impaired Paneth cell function contributes to reduced cell division in the small intestine through its effects on the stem cell niche . It is interesting to note that Zip4 expression in the mouse intestine is restricted to villus enterocytes and this protein is not detected by IHC in the crypts [15] , [18] . Although this cannot exclude the possibility that weak expression of Zip4 in the crypts may contribute to the changes noted in Paneth cells after this gene is knocked out , a more plausible explanation is that loss-of-function of villus enterocyte ZIP4 leads to loss of Paneth cell zinc and it is this loss of zinc which initiates the reprogramming of Paneth cells rather than a loss of a direct signal transduction function of ZIP4 in Paneth cells . The blood circulation provides an intimate connection between the villus and crypt such that enterocyte absorbed nutrients , including Zn , are available at high concentrations to cells in the crypt . There is evidence for zinc transporters affecting cell signaling . Studies of knockout mice suggest that Slc39a14 ( ZIP 14 ) represses phosphodiesterase activity leading to enhanced systemic growth , but the loss-of-function of ZIP14 is not lethal and has only modest effects on growth [50] , unlike the loss of ZIP4 function . Similarly Slc39a13 ( ZIP 13 ) has been shown to play a role in connective tissue development where it affects BMP/TGF-β signaling [51] . The preponderance of data suggests that zinc mediates the effects of these zinc transporters [9] . However , ZnT1 has been shown to directly interact with EVER proteins [52] leading to altered intracellular distribution of zinc , and ZnT1 can interact with Raf-1 kinase through its c-terminal domain leading to activation of ERK signaling [53] . No binding partners of ZIP4 have been reported . In contrast to the pattern of Zip4 expression in villi , Zip5 expression is most active in crypt cells where it localizes to the basolateral membrane when dietary zinc is adequate [15] . This protein is normally degraded during periods of zinc deficiency but the mRNA stays associated with polysomes [12] and its translation is regulated by a conserved stem-loop structure in the 5′-untranslated region and two microRNAs in response to zinc [16] . The reduction of Zip5 mRNA abundance after knocking out the enterocyte Zip4 gene occurs concomitantly with crypt dysplasia . However , intestine-specific knockout of mouse Zip5 is asymptomatic ( Geiser and Andrews , unpublished results ) which suggests that loss of Zip5 expression in the crypts of the Zip4-intestine knockout mice reflects a loss of the differentiated state of intestinal crypt cells . Another rapid effect of loss-of-function of intestine ZIP4 is the attenuation of mTOR1 activity , as measured by phosphorylation of S6 ribosomal protein , in the villus epithelium . This indicates that protein synthesis in villus enterocytes is rapidly diminished after ZIP4 is lost [31] . The intestine has a high rate of protein synthesis to supply digestive enzymes and transporters necessary for enterocyte nutritional functions , and total intestine utilization of amino acids can account for half of whole-body utilization [54] . Attenuation of protein synthesis in the intestine impairs growth via malnutrition . For example , the loss-of-function of intestine Slc6a19 , the neutral amino acid transporter , impairs growth and body weight control in mice and reduces mTOR1 activity in the intestine [55] but is not lethal . Whether loss-of-function of ZIP4 attenuates amino acid uptake by the small intestine remains to be determined but amino acids and growth factors play essential roles in mTOR1 signaling [31] and amino acid stimulation of mTOR1 signaling and protein synthesis are required for intestinal cell migration [33] . The attenuation of mTOR1 activity in villus enterocytes likely contributes to the disorganization of the epithelium which occurs after Zip4 is knocked out . Dietary zinc enhances mTOR1 signaling and GSK-3β phosphorylation in skeletal muscle and liver leading to increased protein synthesis [49] . Similar to Paneth cells , villous enterocytes show altered morphology when intestine Zip4 is knocked out . Combined with the reduced rate of cell division and reduced mTOR1 activity , the loss of normal columnar morphology of enterocytes is consistent with a loss of differentiated function , which likely is a major contributor to growth failure in these mice . Our studies revealed that loss-of-function of intestine ZIP4 leads to transient activation of hepatic AMPK , a cellular energy sensor which is activated when AMP levels increase [32] and which antagonizes mTOR1 activity [32] . Loss of zinc likely leads to attenuated mTOR1 signaling in the liver through reduced production of ATP and activation of AMPK . This is followed by repletion of hepatic zinc due to the redistribution of zinc , and other nutrients , from bone and muscle stores as the knockout mice begin to lose body mass . This leads to attenuation of AMPK activity and increased mTOR1 activity . By extension , the attenuation of villous mTOR1 activity in response to the loss-of-function of ZIP4 in villous enterocytes likely reflects a response to diminished zinc uptake rather than a signaling function of the ZIP4 protein in enterocytes . In addition to the potential reduction in intestinal uptake of amino acids and zinc which can control mTOR1 activity , it is also possible that loss-of-function of ZIP4 attenuates IGF-1 signaling in the intestine . Intestine IGF-1 mRNA is induced after Zip4 is knocked out and this growth factor is a potent tropic mitogen in the intestine [22] . IGF-1 mRNA is induced in intestinal subepithelial myofibroblasts in response to stimulation by glucagon-like-peptide 2 secreted from intestinal L-cells , and IGF-1 enhances crypt cell proliferation by stimulating β-catenin signaling in non-Paneth cells [22] . The finding that IGF-1 is induced but mTOR1 activity is repressed in the villus epithelium suggests that IGF-1 signaling is attenuated when intestine ZIP4 is lost . Intracellular zinc concentrations have been shown to modulate IGF-1 signaling by affecting protein tyrosine phosphatase activity [56] . High zinc is inhibitory leading to enhanced IGF-1 signaling whereas zinc deficiency impairs IGF-1 signaling [57] . Loss of mTOR1 activity in the villus epithelium occurred soon after Zip4 was knocked out in the intestine . In contrast , the activity of mTOR1 was increased in crypts thereafter but this failed to lead to enhanced cell proliferation . The activation of mTOR1 in crypts may reflect a stress response or a compensatory response to crypt dysplasia in the intestine [32] . Increased IGF-1 signaling may lead to activation of mTOR1 in crypt cells causing enhanced translation but fails to enhance cell division due to impaired β-catenin signaling . Clearly a complicated interplay between loss of zinc and altered signal transduction cascades occurs in the intestine leading to diminished integrity of the small intestine in AE . Loss-of-function of intestine ZIP4 led to altered homeostasis of several essential metals over time . Although zinc was rapidly lost from intestine , pancreas and liver soon after the Zip4 knockout , normal levels of zinc re-accumulated in the liver as body weight , bone and muscle mass were lost . In contrast , iron and manganese accumulated in the liver . Accumulation of iron in the liver during zinc deficiency has been previously reported but manganese accumulation has not . Zinc deficiency can result in alterations in iron transporter , storage , and regulatory proteins which facilitate iron accumulation [20] . In addition ZIP14 ( Slc39a14 ) has been suggested to play a role in the accumulation of non-transferrin bound iron in the liver [58] . It is possible that reduced expression of ATP13A2 , which has been shown to reduce cellular manganese concentrations , occurs in response to severe zinc deficiency in our model [59] , and a recent study revealed that ZnT10 ( Slc30a10 ) plays a key role in manganese transport in humans [60] , [61] . Whether enhanced expression or activity of ZIP14 and/or repression of ZnT10 play roles in the accumulation of iron and manganese , respectively , in the liver in response to loss-of-function of intestine ZIP4 remains to be determined . Accumulation of very high levels of iron in the liver ( 20 to 100× ) can lead to hemochromatosis and can inhibit manganese uptake by mitochondria leading to mitochondrial dysfunction [62] . Therefore , the slow accumulation of hepatic iron and manganese in the Zip4-intestine knockout mice could eventually contribute to liver injury if very high levels are accumulated later during the disease process . Perhaps humans with AE may be at risk for hemochromatosis at advanced stages of the disease . In summary , our studies report the first conditional knockout of a member of the mammalian ZIP family of zinc transporters . By controlling the timing and extent of the knockout of the intestine Zip4 gene in mice we were able for the first time to monitor temporal and tissue-specific effects of the loss-of-function of ZIP4 and provide insights into the etiology of AE . The results suggest that Paneth cell dysfunction leading to disruption of the stem cell niche and loss of villus enterocyte integrity are primary causes of AE . These changes likely reflect diminished zinc uptake by ZIP4 and are followed by a switch from anabolic to catabolic metabolism in the mouse leading to dramatic weight loss , dyshomeostasis of several essential metals and ultimately lethality in the absence of excess dietary zinc .
Experiments involving mice were performed in accordance with the guidelines from the National Institutes of Health for the care and use of animals and were approved by the Institutional Animal Care and Use Committee . Mice were maintained on normal mouse chow ( zinc adequate:ZnA ) and where indicated were also provided with drinking water containing 250 ppm ZnSO4 ( zinc excess: ZnE ) . We previously described the structure of the mouse Zip4 ( Slc39a4 ) gene in detail [15] , [18] . The final structure of the floxed Zip4 gene targeting vector created herein and of the targeted chromosomal locus is shown in Figure 1 . BAC recombineering [63] using galK selection was employed to manipulate the Zip4 gene [64] A 9180 bp region containing the Zip4 gene plus 2 . 3 kb of 5′-flanking DNA and 3 . 2 kb of 3′ flanking DNA was captured from BAC ct7-43303 and gap-repaired in a conditionally amplifiable [65] BAC-based vector ( P[acman]-M-KO ) [66] that allows for negative selection ( HSV-TK ) in ES cells and positive selection in bacteria ( ampicillin ) . A LoxP site engineered to contain an Xho I restriction site was inserted into a poorly conserved region in intron 5 and a second LoxP site was inserted immediately downstream of exon 12 , the last exon in this gene . An mc1-neomycin cassette was inserted about 250 bp downstream of exon 12 . The targeting vector's structure was confirmed for the entire gene including the neomycin cassette by DNA sequencing and the ends of the captured regions were sequenced to verify that no rearrangements or mutations had occurred . In addition , the functionality of the LoxP sites was confirmed by transformation of the final targeting vector into bacteria ( EL 350 ) that express Cre recombinase under control of an arabinose inducible promoter [63] . The Transgenic and Gene-Targeting Institutional Facility at the KU Medical Center generated targeted ES cell clones and performed blastocyst injections . The Zip4 FX targeting vector was linearized with Not1 and electroporated into E14 embryonic stem ( ES ) cells . Selected colonies were screened by long range PCR using LA-Taq ( TaKaRa Bio , Inc . ) and primers which flank the Zip4 region captured and manipulated by recombineering paired with primers within the Zip4 gene itself ( Figure 1 ) . Primers flanking the LoxP site in intron 5 were used for routine genotyping of the targeted allele in mice ( Figure 1 ) . Homologous recombination of the targeting vector into the endogenous locus resulted in the insertion of an XhoI site engineered into the LoxP sequence in intron 5 which aided in identifying the targeted alleles . Southern blot hybridization was used to screen for the Y-chromosome as described in detail previously [67] . The sequences of oligonucleotides for integration screen and genotyping are shown in Table S1 . Chimeric mice were generated by microinjection of two independent Zip4FX/WT ES cell clones into d4 C57BL/6 blastocysts , followed by transfer to pseudopregnant CD-1 foster mothers . Resulting chimeric mice were crossed with C57BL/6 females ( Harlan labs ) . Germline transmission was confirmed by PCR from tail DNA of agouti offspring ( Figure 1 ) . Zip4FX/WT mice were crossed and Zip4FX/FX offspring were identified by PCR amplification of the LoxP:XhoI insertion in intron 5 . Zip4FX/FX mice were crossed to create a working colony of mice homozygous for floxed Zip4 genes . These mice were then crossed with transgenic mice bearing a tamoxifen-dependent Cre recombinase ( vil-Cre-ERT2: a kind gift from S . Robine , Institut Curie-CNRS , Paris , France ) expressed under the control of the villin promoter to allow for inducible deletion of the Zip4 gene specifically in the intestinal epithelium [11] and selected offspring were backcrossed to yield mice heterozygous for vil-Cre-ERT2 and homozygous Zip4FX/FX . These mice were then crossed with Zip4FX/FX mice to yield 50% offspring with Zip4FX/FX: vil-Cre-ERT2 alleles and 50% with Zip4FX/FX alleles . The latter provided age and genetically matched controls for our experiments and were labeled as control ( Con ) in all the figures . A tamoxifen stock solution was prepared essentially as described previously [68] by the addition of 100 µl of ethanol to 10 mg of tamoxifen ( free base: MP biomedicals , LLC ) and heating to 37°C briefly to dissolve the tamoxifen . This solution was then diluted to 1 ml with autoclaved canola oil and heated briefly to 37°C . The stock solution ( 10 mg tamoxifen/ml ) was stored at 4°C for up to two weeks or at −80°C for months . Before injection , the tamoxifen stock solution was heated to 37°C . Adult and recently weaned mice ( 5 to 8 days post-weaning ) were injected I . P . with 100 µl ( 1 mg tamoxifen ) of the tamoxifen stock solution daily for 3 consecutive days unless noted otherwise ( e . g . Figure 2F ) . Neonatal mice ( d5 post-partum ) were injected I . P . for 5 consecutive days with 10 µl ( 100 µg ) of the tamoxifen stock solution . Total RNA was isolated using Trizol reagent ( Invitrogen ) . Total RNA ( 3–6 µg ) was size-fractionated by agarose-formaldehyde gel electrophoresis , transferred and UV cross-linked to a Zeta Probe GT nylon membrane ( BioRad ) . Northern blot membranes were prehybridized in 0 . 5 M sodium phosphate , pH 7 . 0 , 7% SDS at 65°C for 30 min and then hybridized in 40% formamide , 0 . 5 M sodium phosphate , pH 7 . 0 , 7% SDS at 65°C for 24 hr . Membranes were then washed twice in 3X SSC , 0 . 1% SDS for 30 min each followed by two washes in 1X SSC , 0 . 1% SDS and two washes in 0 . 3X SSC , 0 . 1% SDS for 30 min each at 65°C . Hybrids were detected by autoradiography at −80°C . Duplicate gels were stained with acridine orange to ensure equivalent loading and integrity of total RNA . Riboprobes for mouse MT-I , Zip4 , and Zip5 [15] and for IGF-1 mRNAs were as described [69] . Probes were used at 2×106 cpm/ml of hybridization solution . Small intestine total RNA was reverse transcribed using Accuscript reverse transcriptase ( RT ) enzyme ( Stratagene ) and IGF mRNA and GAPDH mRNA , as an internal control , were amplified using the PerfeCTa Sybr Green FastMix qPCR kit ( Quanta Biosciences ) and a Miniopticon Real-Time PCR Detection System ( BioRad ) essentially as described previously [8] . Melting-curve data were collected to confirm PCR specificity . Each cDNA sample was run in triplicate , and the corresponding no-RT sample was included as a negative control . GAPDH primers were included in every sample to control for sample variation . IGF-1 mRNA product in each sample was normalized to that of the GAPDH mRNA . The amount of PCR products was measured by threshold cycle ( Ct ) values and relative mRNA levels were determined as unit values of 2∧[Ct ( GAPDH ) -Ct ( IGF-1 ) ] . Values presented are fold-change relative to control . The following primers were used: mIGF-1 ( s ) gctcttcagttcgtgtgtggaccg: mIGF-1 ( as ) cttctgagtcttgggcatgtcagtgtg . These primers amplify the IGF-1 peptide sequence found in all mIGF-1 variants . GAPDH ( s ) : aggttgtctcctgcgacttca: mGAPDH ( as ) ccaggaaatgagcttgacaaag . The following antibodies were used at the indicated dilution for Western blots: Total β-catenin ( Cell Signaling: 1∶1000 ) , active β-catenin ( Millipore: 1∶1000 ) , p ( T172 ) -AMPK ( Cell Signaling: 1∶1000 ) , total AMPK ( Cell Signaling: 1∶1000 ) , p ( S2448 ) mTOR1 ( Cell Signaling: 1∶1000 ) , total mTOR1 ( Cell Signaling: 1∶1000 ) , p ( s473 ) -AKT ( Cell Signaling: 1∶1000 ) , total AKT ( Cell Signaling: 1∶1000 ) . The following antibodies were used at the indicated dilution for immunohistochemistry ( IHC ) : Sox9 ( Millipore: 1∶750 ) , lysozyme ( DakoCytomation: 1∶150 ) , phospho-S6 ribosomal protein ( Cell Signaling: 1∶400 ) , phosphor ( Ser10 ) -histone H3 , ( Cell Signaling: 1∶100 ) The first ∼5 cm of small intestine ( 3 to 5 mice per group ) was collected , flushed with cold PBS , cut into small pieces and fixed in Bouin's fixative or 4% paraformaldehyde in PBS overnight at 4°C . Fixed tissues were embedded in paraffin and sections ( 1 µm ) were prepared by Histo-Scientific Research Laboratories ( HSRL ) . Bouin's fixed sections were deparaffinized , rehydrated and stained with hematoxylin-eosin for examination of gross morphology . Paraformaldehyde fixed tissues were deparaffinized , rehydrated and stained using the Periodic Acid-Schiff ( PAS ) staining system ( Sigma-Aldrich ) to detect mucin ( mainly Muc2 and Muc3 ) and identify goblet cells . Serial sections were pretreated with α-amylase ( 239 µg/ml for 1 hr at 37°C: Worthington Biochemical Corporation ) to test for glycogen contribution to the PAS staining . For IHC , paraformaldehyde fixed serial sections were deparaffinized , rehydrated and antigens were retrieved in 10 mM citrate , pH 6 . 0 , using a 2100 Retriever pressure cooker ( PickCell Laboratories ) . Proteinase K digestion ( 20 µg/ml in 50 mM Tris base , pH 8 . 0 , 0 . 1 mM EDTA , 0 . 5% Triton X- 100 at 37°C for 10 min ) was used for antigen retrieval for the lysozyme antibody . Samples were processed using the Histostain Plus LAB SA Detection System ( Invitrogen ) according to the manufacturer's instructions using the antibodies listed above . Stained slides were counterstained briefly in Mayer's hematoxylin ( Sigma ) and photographed using a Leica DM 4000B microscope ( Leica-microsystems ) with Adobe Photoshop image capture software ( Adobe ) . Paneth cell labile zinc was detected using the zinc-specific-responsive fluorescent dye Zinpyr-1 ( ZP1 ) ( Millitech ) essentially as described with modifications [26] . The small intestine from 3 to 5 mice per group was isolated , flushed with cold PBS and cut into small pieces that were dropped into cold 0 . 6 M sucrose in PBS . Once the small intestine pieces sank ( 1 to 2 min ) they were transferred to a solution of 2 parts 0 . 6 M sucrose plus 1 part TFM embedding compound ( Triangle Biomedical Sciences , Inc . ) until they sank ( 2 to 3 min ) . The intestine pieces were then immersed in TFM , frozen in dry ice-propanol and 6 µm frozen sections were cut . Frozen sections were thawed for ∼30 seconds at room temperature in a solution containing 20 µM ZP1 and 0 . 5 µM DAPI in 147 mM NaCl , 4 mM KCl , 3 mM CaCl2 , 0 . 9 mM MgCl2 , 11 mM HEPES , pH 7 . 4 , and 10 mM glucose , washed briefly in PBS and then photographed using a Leica DM 4000B microscope ( Leica-microsystems ) with Adobe Photoshop image capture software ( Adobe ) . ZP1 was visualized using the +L5 ( Leica-microsystems ) GFP filter cube and DAPI stained nuclei were visualized using the +A4 filter cube ( Leica-microsystems ) . Brightness of the photographs was reduced in Photoshop . Liver proteins were isolated by homogenizing ∼100 mg of tissue in 400 µl of RIPA buffer ( 150 mM NaCl , 50 mM Tris base , pH 8 . 0 , 1% NP40 , 0 . 5% deoxycholate , 0 . 1% SDS , plus Complete Protease Inhibitor Cocktail and PhosStop phosphatase inhibitor cocktail ( Roche Applied Science ) . Tissues were homogenized at 4°C for 20 strokes using a teflon-glass homogenizer . Samples were then centrifuged at 10 , 000× g for 10 min at 4°C , the supernatant was collected and protein concentration was determined using BCA reagent ( Pierce Biotech ) . Proteins ( 30 µg ) were resolved on 7 . 5% or 10% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes ( Life Science Products , Inc . ) using a Transblot semi-dry transfer apparatus ( BioRad ) according to the manufacturer's instructions . Membranes were blocked overnight in 5% milk in PBS-T or 5% BSA in PBS-T as suggested by the antibody source and then incubated with primary antibody in blocking solution at the appropriate dilution for 2 h at room temperature . After extensive washing , membranes were incubated with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody ( 1∶2000 ) and the blot was developed using ECL Plus reagent ( Amersham Biosciences ) according to manufacturer's instructions and detected using Kodak BioMax MS film ( Kodak ) . Elemental profiling via inductively coupled plasma mass spectrometry ( ICP-MS ) was performed at the Purdue University Ionomics Facility , Purdue University , West Lafayette , Indiana as described previously [21] . The following elements were measured: Na , Mg , P , K , Ca , Fe , Co , Cu , Zn , Mn , As , Se , and Mo . Mouse tissues ( n = 5 to 10: <100 mg wet weight each sample ) were dried at 95°C in a vacuum oven . Dried samples of about 5 to 10 mg were weighed into Pyrex tubes and digested in 0 . 7 to 1 . 5 ml concentrated HNO3 at 110°C for 4 h . Each sample was diluted as optimal with 18 MOhm water and analyzed on an Elan DRCe ICP-MS ( PerkinElmer ) . Methane was used as a collision cell gas to measure iron . Gallium and indium were used as internal standards , added to the digestion acid bottle to a concentration of 20 µg l−1 . National Institute of Standards and Technology traceable single element ICP standards ( Ultrasci ) were used to make up the calibration standards . Tissue concentrations were determined as µg g−1 dry weight ( ppm ) of each element . Graphs were generated and statistical analyses performed using GraphPad Prism5 software ( GraphPad Software ) . Statistical significance was determined using the Unpaired T-test ( two-tailed ) and values were considered different if P<0 . 05 . Data are expressed as the mean ± S . E . M . *indicates P<0 . 05; *** indicates P<0 . 001; **** indicates P<0 . 0001 | Loss-of-function of the zinc transporter ZIP4 in the mouse intestine mimics the lethal human disease acrodermatitis enteropathica . This is a rare disease in humans that is not well understood . Our studies demonstrate the paramount importance of ZIP4 in the intestine in this disease and reveal that a root cause of lethality is disruption of the intestine stem cell niche and impaired function of the small intestine . This , in turn , leads to dramatic weight loss and death unless treated with exogenous zinc . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"small",
"intestine",
"cellular",
"stress",
"responses",
"animal",
"genetics",
"genetic",
"mutation",
"anatomy",
"and",
"physiology",
"gene",
"function",
"animal",
"models",
"genetics",
"of",
"disease",
"physiological",
"processes",
"immunochemistry",
"model",
"organisms",
"nutrition",
"homeostasis",
"gastroenterology",
"and",
"hepatology",
"biological",
"transport",
"dna",
"digestive",
"system",
"metabolic",
"pathways",
"biology",
"mouse",
"molecular",
"biology",
"energy",
"metabolism",
"metabolic",
"disorders",
"mutagenesis",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"physiology",
"genetics",
"digestive",
"functions",
"dna",
"recombination",
"molecular",
"cell",
"biology",
"metabolism",
"digestive",
"physiology",
"genetics",
"and",
"genomics",
"micronutrient",
"deficiencies"
] | 2012 | A Mouse Model of Acrodermatitis Enteropathica: Loss of Intestine Zinc Transporter ZIP4 (Slc39a4) Disrupts the Stem Cell Niche and Intestine Integrity |
Repetitive DNA elements are mutational hotspots in the genome , and their instability is linked to various neurological disorders and cancers . Although it is known that expanded trinucleotide repeats can interfere with DNA replication and repair , the cellular response to these events has not been characterized . Here , we demonstrate that an expanded CAG/CTG repeat elicits a DNA damage checkpoint response in budding yeast . Using microcolony and single cell pedigree analysis , we found that cells carrying an expanded CAG repeat frequently experience protracted cell division cycles , persistent arrests , and morphological abnormalities . These phenotypes were further exacerbated by mutations in DSB repair pathways , including homologous recombination and end joining , implicating a DNA damage response . Cell cycle analysis confirmed repeat-dependent S phase delays and G2/M arrests . Furthermore , we demonstrate that the above phenotypes are due to the activation of the DNA damage checkpoint , since expanded CAG repeats induced the phosphorylation of the Rad53 checkpoint kinase in a rad52Δ recombination deficient mutant . Interestingly , cells mutated for the MRX complex ( Mre11-Rad50-Xrs2 ) , a central component of DSB repair which is required to repair breaks at CAG repeats , failed to elicit repeat-specific arrests , morphological defects , or Rad53 phosphorylation . We therefore conclude that damage at expanded CAG/CTG repeats is likely sensed by the MRX complex , leading to a checkpoint response . Finally , we show that repeat expansions preferentially occur in cells experiencing growth delays . Activation of DNA damage checkpoints in repeat-containing cells could contribute to the tissue degeneration observed in trinucleotide repeat expansion diseases .
Repetitive DNA is found dispersed throughout eukaryotic genomes , and in some cases is central to key biological processes such as chromosome segregation and chromosome end protection [1] . Repeat tracts are usually sites of variation among individuals , with some classes of repeats expanding to sizes that cause pathology . For example , expansion of CAG/CTG trinucleotide repeats ( abbreviated CAG ) have been observed to occur at several different genomic loci , causing diseases that include Huntington's disease , myotonic dystrophy , and multiple subtypes of spinal cerebellar ataxia [2]-[3] . CAG trinucleotide repeats are among a class of repeats that are unique in that they form hairpin secondary structures that interfere with DNA replication and DNA repair [1] , [4] . The repeats exhibit a threshold length beyond which expansions become increasingly likely . For CAG repeats in humans , the expansion threshold is 35–38 repeats , 100–115 bp . In addition to the instability threshold , a disease-causing threshold also exists for trinucleotide repeats , which is at or above the expansion threshold , and is dependent on the locus and disease process . For Huntington's disease the disease-causing threshold is 38–40 repeats , and is governed by the length at which the polyglutamine tract ( coded for by CAG ) within the Huntingtin gene becomes toxic . At the myotonic dystrophy locus , the disease threshold is closer to 200 repeats , the size at which the CUG RNA exerts toxic effects on muscle cells [2] , [4] . It is well established that in mammalian cells , proteins with an abnormally long polyglutamine tract due to a CAG expansion cause toxic effects that ultimately result in cell death [2]-[3] . In addition , RNA containing a long CUG tract can also cause toxicity and cell death by sequestering RNA binding proteins , as happens in patients with myotonic dystrophy where the expanded CTG repeat is transcribed but not translated [2]-[3] . However , less is known about whether the expanded repeat DNA itself is toxic to cells . CAG repeats of 55–80 repeats have been shown to block replication fork progression in plasmids , cause fork reversal on a eukaryotic chromosome [5]-[7] , and interfere with ligation of 5′ DNA flaps that occur during Okazaki fragment maturation or gap repair [8]-[9] . Structure-forming trinucleotide repeats also cause double-strand breaks ( DSBs ) in a length-dependent manner resulting in chromosome fragility [10]-[11] . Thus multiple types of DNA damage occur at an expanded trinucleotide repeat tract , including stalled or reversed forks , single-strand breaks or gaps , and double-strand breaks ( DSBs ) . Our recent results have shown that DSBs at expanded CAG tracts are healed by both Rad52-dependent homologous recombination ( HR ) and Dnl4 ( Lig4 ) mediated end joining ( EJ ) pathways , and that integrity of the MRX complex ( Mre11/Rad50/Xrs2 ) is critical for preventing chromosome fragility at CAG repeats [12] . Given the evidence that these structure-forming repetitive DNA elements are damage-prone , it is critical to determine whether they are capable of eliciting cell cycle checkpoint responses independent of defects in RNA or protein metabolism . Sites of DNA damage recruit proteins directly involved in DNA repair or fork restart as well as checkpoint proteins [13]-[14] , and the persistence of DSBs or stalled forks in mutants deficient for key checkpoint factors suggests that repair mechanisms are aided by cell cycle checkpoint activation [15]-[16] . Recruitment and activation of a checkpoint protein triggers a signaling cascade mediated by substrate phosphorylation , leading to downstream responses such as halting cell cycle progression to allow time for repair . The Rad53 ( Chk2 ) checkpoint kinase is the central effector of the DNA damage checkpoint in S . cerevisiae [16]-[17] . Activation of Rad53 in S phase leads to inhibition of late origin firing , stabilization of stalled replisomes , and a slowing of S phase progression in what is referred to as the S phase checkpoint , although there is no actual S phase arrest and cells will eventually enter G2 [18]-[19] . Not all checkpoint activation leads to a noticeable cell cycle delay , especially when the damage is repaired quickly [20] . However if the damage is not repaired in a timely manner , an extended checkpoint arrest will ensue , and cells will accumulate at the G2/M boundary . In yeast , a cell harboring a single , irreparable DSB can adapt to the damage after 8–10 hours and continue cycling; the damage may be then be repaired in a subsequent cycle [21]-[23] . In both yeast and mammalian cells , persistence of the damage will eventually result in cell death [24]-[25] . Thus induction of a DNA damage checkpoint can either facilitate repair or lead to cell death , depending on the severity of the damage . The DNA damage checkpoint has been primarily studied as a response to exogenous damage to the genome by agents that cause fork stalling , base damage , or DSBs . However less is known about the response to endogenous sequences with the potential to form non-B DNA structures . Previous genetic evidence in S . cerevisiae indicated that checkpoint proteins were involved in CAG repeat tract maintenance , suggesting that expanded repeats accumulate damage that is sensed by the cell cycle checkpoint machineries . Mutation of genes central to the DNA replication and DNA damage checkpoints including MEC1 , MRC1 , RAD24 , RAD17 , and RAD53 resulted in elevated rates of repeat-mediated chromosome fragility and instability [26]-[28] . Particularly , deletion of genes that sense and transduce the replication checkpoint response , MRC1 , MEC1 , and RAD53 , led to the greatest increase in CAG fragility [26]-[27] . Although these checkpoint proteins are crucial for preventing or healing CAG breaks , direct evidence for whether the expanded repeats activate cell cycle arrests , if so at what frequency , and any consequence for cell growth and viability , was lacking . Using cell cycle , biochemical and functional assays that can measure the robustness of checkpoint activation in repeat-containing cells , we report the novel observation that damage at structure-forming ( CAG ) 70 or ( CAG ) 155 repeats has the potential to activate DNA damage checkpoints , interfere with normal cell cycle progression , and compromise cell proliferation and survival in S . cerevisiae .
Our earlier results indicated that checkpoint proteins are required in cells containing expanded CAG repeats in order to prevent increased fragility and instability of the repeats [26]-[27] . However , as these results were partly obtained using genetic assays that could detect rare events , it was not known if checkpoint activation in repeat-containing cells is a rare or common event , or whether it results in cell cycle delays . Therefore we sought to determine whether an expanded CAG repeat causes measurable effects on cell growth . We used a yeast strain containing a yeast artificial chromosome ( YAC ) with either no repeat ( CAG-0 ) or an expanded CAG repeat originally cloned from a myotonic dystrophy patient [29] . The repeat was cloned into a region of the YAC not predicted to be transcribed or translated , and flanked by minimal human sequence , 50 bp and 150 bp on each side the repeat . In general , two allele sizes were studied , 70 and 155 repeats , which have been previously shown to exhibit both instability and fragility at this location [29] . Preliminary experiments in liquid culture revealed slight delays in exponential growth of wild-type yeast containing a CAG-195 repeat at this location ( Figure S1 ) . However , because bacterial cells with contracted repeats have a growth advantage in liquid culture over those with longer repeats [30] , and since about 30% of CAG-155 repeats have a contraction event in wild-type ( WT ) yeast [29] , we reasoned that the observed repeat-induced growth inhibition could be an underestimation . Therefore , a microcolony experiment , which measures population viability of cells on solid media , was performed to assess the extent of repeat-mediated growth inhibition . This experiment is similar in principle to that performed by Weinert and Hartwell [31] , which originally described mutants that escaped the control of DNA damage checkpoints in S . cerevisiae . Unbudded G1 cells were micromanipulated onto YC-Leu solid media that maintained selection for the repeat-containing chromosome , and their growth into microcolonies ( small colonies , visible under the microscope ) was monitored ( Figure 1A ) . Since no essential genes required for cell survival are located in close proximity to the repeat , the growth differences observed between cells with or without the CAG tract can be attributed to a repeat-mediated effect . A typical S . cerevisiae cell cycle time of 2 hours on synthetic media would result in about 15 doublings at 30 hours of growth . At 30 hours , we observed a bimodal distribution of microcolony growth in WT CAG-70 or CAG-155 strains . This distribution typically consisted of “survivors” ( area≥0 . 01 mm2 , more than 6 doublings , Figure 1A−1E ) and “non-survivors” ( area≤0 . 005 mm2 , 6 or fewer doublings ) . The non-survivors likely represent terminally arrested lineages ( frequencies , reported in [12] , were ∼10% for WT cells with or without a CAG repeat ) . Among survivors , WT CAG-70 or CAG-155 strains showed a significant 1 . 5 to 2-fold decrease in mean microcolony area compared to the WT CAG-0 control strain ( Figure 1B ) . Therefore , in the WT strain , a large enough proportion of the cells containing an expanded CAG repeat experienced a growth delay that the overall rate of population growth was affected . We have seen the same effect in two other yeast strain backgrounds , ( A . LaPorte , C . Weindel , and C . H . Freudenreich , data not shown ) . We also assessed the survival efficiency of mutants lacking end-joining ( dnl4Δ ) , homologous recombination ( rad52Δ ) or the MRX complex ( mre11Δ ) because we showed that all three of these pathways act to repair breaks at expanded CAG-70 or -155 repeats [12] . The dnl4Δ strain exhibited a pattern similar to WT , with CAG-70 and CAG-155 microcolonies attaining a mean area 3 . 3-fold and 1 . 2-fold below the CAG-0 control respectively ( Figure 1C ) . The rad52Δ CAG-70 and CAG-155 microcolonies attained mean sizes 5 . 2 and 3 . 6-fold below CAG-0 , revealing that the presence of expanded repeats significantly inhibited population growth in this background and more severely than in the WT strain ( Figure 1D ) . Overall , the extent of repeat-induced growth inhibition depended on both the length of the repeat and the nature of the repair defect: Rad52 deficiency severely impacted the viability of both CAG-70 and CAG-155 survivors , while the Dnl4-EJ pathway deficiency impacted the viability of cells with 70 repeats more than those with 155 repeats , a trend that was present in all three backgrounds but was more pronounced in the dnl4Δ mutant . Surprisingly , the repeat-induced growth defect phenotype observed in the rad52Δ and dnl4Δ repair mutants was absent among mre11Δ survivors ( Figure 1E ) . Deletion of MRE11 caused a significant growth defect in all cells , with or without an expanded CAG repeat . The broader area distribution and relatively smaller colony sizes of both rad52Δ CAG-0 and mre11Δ CAG-0 strains indicate that a wide spectrum of sporadically occurring damage in addition to CAG repeat damage require HR and MRX pathways for viability . However , the slow growth phenotype of the mre11Δ CAG-70 strain , rather than being enhanced , was somewhat relieved relative to CAG-0 or CAG-155 strains . This striking phenotype suggests that a proportion of damage that occurs at a CAG-70 repeat can escape detection by the cell cycle checkpoint machinery in the absence of the MRX complex . In addition to the effects graphed in Figure 1 , a significant 3 . 5-fold increase in the frequency of non-survivors was observed in strains with CAG-70 or -155 repeats compared to the CAG-0 control in the rad52Δ and mre11Δ backgrounds , an effect not observed in wild-type or dnl4Δ strains [12] . Collectively , the above phenotypes indicate that repeat-containing cells have a proliferation defect frequent enough to result in reduced population viability , and that deficiency of DSB repair pathways exacerbates the defect . To better understand the basis of smaller microcolony sizes and reduced population viability of repeat-containing cells , a single cell pedigree analysis was performed . Single precursor cells were micromanipulated onto YC-Leu solid media , and successive divisions were followed for a span of ∼18 hours , micromanipulating the daughter cells away at each division in order to follow a single lineage ( Figure 2A ) . For each strain , about 30 lineages were analyzed for 5–7 divisions per lineage , for an average of 90 cell divisions per strain ( Table S1 ) . We first analyzed the division potential of each lineage , i . e . the ability of a progenitor cell with an expanded CAG repeat to sustain successive divisions in a lineage ( Figure 2B ) . In the wild-type background , the CAG-70 and CAG-155 repeat-containing strains completed on average fewer divisions per lineage ( 3 . 6 and 4 . 1 ) than the CAG-0 no-tract control strain ( 4 . 6 ) . This difference was even more striking in dnl4Δ and rad52Δ backgrounds , with fewer divisions completed compared to the no repeat control and some lineages failing to complete even one division ( e . g . 60% for rad52Δ CAG-155 , Figure 2B ) . Similar to the microcolony analysis , the profile was altered in the mre11Δ strain , with CAG-70 cells exhibiting a greater division potential than either CAG-0 or CAG-155 strains . This result supports the conclusion that damage at the CAG-70 tract is not being properly sensed or signaled in the absence of the MRX complex . This conclusion is further substantiated by a decreased frequency of cell-cycle arrests and decreased morphological defects in mre11Δ CAG-70 cells compared to mre11Δ CAG-0 and CAG-155 cells ( see below , Figure 2C and Figure 3B ) . To determine if reduced division potential was due to frequent cell cycle arrests , divisions were either categorized as normal divisions , lasting <3 . 0 hours , or arrest divisions , lasting >3 . 0 hours . While a majority of divisions in the WT CAG-70 and CAG-155 strains were normal divisions ( cycling time of <3 hours ) , 55% and 38% of divisions , respectively , were >3 hour arrest divisions , an elevated frequency compared to the WT CAG-0 control ( 29%; Figure 2C ) . Both dnl4Δ and rad52Δ strains showed a further disparity between repeat-containing and no-tract control strains , with a 2 . 6-fold increase for CAG-70 and 3 . 3 to 3 . 5-fold increases for CAG-155 strains in the frequency of >3 hour divisions ( Figure 2C ) . For the rad52Δ CAG-155 strain , about 60% of divisions took longer than 3 hours , compared to only 17% in the rad52Δ CAG-0 control . Thus the presence of an expanded CAG repeat had the potential to prolong cell division in a large proportion of cells , and arrests were 2 to 3-fold more frequent than in cells without an expanded repeat tract . The increased frequency of arrests explains the repeat-specific decrease in microcolony size and reduced division potential . In addition to “normal” arrest times of 4–6 hours , we noted that a subset of arrests were very prolonged , >8 hours and up to 16 or more hours ( the length of the experiment ) ( Figure 2C and Figure S2 ) . Certain repair foci are known to last >8 hours in arrested cells , although yeast cells can also sometimes adapt after ∼8 hours of arrest and re-enter the cell cycle even though the damage has not been repaired [21] , [32]-[33] . Adaptation coincides with release from arrest , re-entry into the cell cycle , and disappearance of Rad53 phosphorylated species [34] . WT CAG-70 and CAG-155 strains experienced >8 hour arrest divisions at a frequency 3 to 5-fold higher than the WT CAG-0 control ( Figure 1C ) . Among the DSB repair-deficient mutants , 6 . 3% or 4 . 4% of dnl4Δ CAG-70 and CAG-155 strains experienced arrest divisions >8 hours , respectively , relative to 0% in the dnl4Δ CAG-0 control . The rad52Δ CAG-70 and CAG-155 strains experienced >8 hour arrests in 7 . 8% and 23% of divisions respectively , an 11- or 33-fold increase over the rad52Δ CAG-0 control . The higher frequency of >8 hour arrest divisions observed in repeat-containing cells suggests that expanded repeats are accumulating damage that is difficult to repair . Finally , among all comparisons , the dramatic increase in the frequency of >8 hour arrest divisions in the rad52Δ CAG-155 strain , which is 6 . 4-fold greater than the WT CAG-155 strain , indicates a requirement for Rad52-mediated HR in repairing damage at the CAG-155 repeat and ensuring timely cell cycle progression . We further analyzed whether the arrests resulted in recovery , where the cell cycle resumed before 8 hours of arrest ( usually indicating successful repair ) , adaptation , where the cell was arrested more than 8 hours but did eventually re-enter the cell cycle , or terminal arrests , which were not observed to re-enter the cell cycle . For wild-type cells , most arrest events were able to recover or adapt ( 96–100% ) , although the recovery frequency was lower for repeat-containing cells ( Table 1 ) . Cells lacking one of the main DSB repair proteins and containing a repeat tract were less likely to recover , and more likely to have an adaptation response or experience a terminal arrest . Notably , with the exception of the mre11Δ mutant , terminal arrests were not observed in the CAG-0 control strain even in rad52Δ or dnl4Δ backgrounds , suggesting that the presence of an expanded repeat can lead to a more severe outcome than deficiency in a major DSB repair pathway . However the most striking difference between cells with and without an expanded repeat was the likelihood of experiencing a second arrest in the next cell division , as these recurring arrests appeared only in repeat-containing lineages ( Table 1 ) . This phenomena was repeat-length dependent , occurring more often in cells with a CAG-155 than CAG-70 repeat , and the second arrest was often longer than the first . Apparently , the damage at a repeat tract can sometimes persist through mitosis , causing a second arrest in the next cell cycle . It is known that yeast cells undergoing a protracted arrest will often exhibit abnormal morphology , including swelling and elongated or multiple buds [35]-[36] . Indeed , we noted these phenotypes were increased in repeat-containing cells compared to the no repeat control ( Figure 3 ) . Several of the trends noted in our other assays were confirmed by comparing the occurrence of abnormal morphology: the occurrence of a swollen or multi-budded phenotype was exacerbated by increasing repeat length , with the Rad52 protein appearing to be particularly important for preventing CAG-155 repeat-induced effects , and the Dnl4 protein being either equally required at the two repeat lengths or slightly more so at CAG-70 . In contrast , for the mre11Δ mutant , the repeat-containing cells had less severe swelling than the no-tract control and multi-budded cells were minimal to absent . This profile indicated that the repeat-induced checkpoint was bypassed in this background , especially in the CAG-70 strain , unlike other non-repeat damage that was sensed and induced a response ( Figure 3B ) . Cytokinesis defects and cell swelling similar to those observed in the repeat-containing cells have been documented in cells exposed to hydroxyurea ( HU ) , which depletes nucleotide pools and results in slowing or stalling of replication forks [36]-[37] . We confirmed that a WT CAG-0 strain treated with a sub-lethal dose of HU exhibited a 2 to 3-fold increase in swollen and multi-budded cells ( Figure S3 ) . These data implicate replication stress as one likely cause of the phenotypes observed in the repeat-containing cells . In summary , our results indicate that expanded CAG repeats have the potential to induce a protracted cell division cycle accompanied by frequent swelling and morphological defects , all of which are phenotypes that occur during activation of DNA damage checkpoints . These phenotypes were exacerbated in backgrounds deficient for either end joining or homologous recombination repair , and in contrast were mitigated by the absence of the Mre11 protein , especially for a CAG-70 repeat tract . In order to determine at what point in the cell cycle delays were occurring , and thereby gain insight into the potential types of damage causing the delays , we determined the cell cycle distribution of growth arrests . The frequencies of G1 , S and G2/M arrests were recorded in wild-type or mutant cells with or without an expanded repeat tract ( Figure 4A ) . Analysis of individual divisions from the pedigree experiment revealed that wild-type repeat-containing cells showed a bias towards arresting in the S and G2/M cell cycle phases relative to a CAG-0 control ( Figure 4A ) . Specifically , the CAG-70 strain exhibited a modest but significant increase in S phase arrests ( to 33% vs . 28% for CAG-0 ) and a greater tendency to arrest in G2 ( 51% vs . 40% ) , whereas the CAG-155 strain exhibited a further increase in frequency of arrests that began in S phase ( 38% ) . The profile was shifted dramatically in a rad52Δ strain , with the vast majority of the arrests occurring in the G2 phase ( 71–93%; absolute percentages are 2 . 8 and 2 . 7-fold over CAG-0 , Table S1 ) . Furthermore , the increase in S phase arrests in cells with a CAG-155 repeat was evident in the rad52Δ background and was highly significant compared to rad52Δ CAG-0 or CAG-70 ( 26% vs . 7–8% , Figure 4A ) . Altogether , these data suggest that the CAG-155 repeat results in a perturbation of replication that is severe enough to cause a slowing of S phase , while the damage at CAG-70 , although it may also originate in S phase , does not always induce the S phase checkpoint but rather is more often resolved in G2 . Furthermore , Rad52-dependent HR is apparently a crucial pathway for repairing repeat-induced DNA damage , a result consistent with the increased repeat tract fragility observed in a rad52Δ strain [12] . Shifts in arrest phase also occurred in repeat-containing dnl4Δ and mre11Δ cells; the majority of arrests in the absence of either protein were at the G2/M boundary , with mre11Δ cells additionally exhibiting a reduction in G1arrest frequencies compared to the CAG-0 control ( Figure S4 ) . To determine whether repeat-specific S and G2 arrests were also visible at a population level , cell cycle distributions were analyzed by flow cytometry ( FACS ) . In this method , all cells are analyzed , not just those undergoing an arrest . The S phase slowdown was more difficult to detect by FACS than single cell analysis , but the tendency of WT repeat-containing cells to accumulate with 2N DNA content ( G2/M ) was evident ( Figure 4B ) . In contrast , the WT CAG-0 strain contained mostly cells with 1N DNA content ( G1/G0 ) near stationary phase . In the rad52Δ background , the proportion of cells that accumulated with 2N DNA content was increased even in the CAG-0 control but was further increased in the CAG-155 strain , confirming the importance of this protein for repair of repeat-induced damage in addition to spontaneous damage occurring within non-repetitive regions . Based on the above phenotypes , we conclude that expanded CAG repeats have the potential to induce both the intra-S and G2/M DNA damage checkpoints . Our results show that specific S and G2/M phase arrests , as opposed to a general overall slowed progression through the cell cycle , contribute to longer cell division times in repeat-containing cells . Rad52-dependent homologous recombination is particularly crucial for prevention of S phase delays and release from the G2/M block to allow resumption of cell proliferation . Since the above results indicated that checkpoint-mediated cell cycle arrests are associated with a repair or fork restart event , we wished to determine the status of the CAG repeat locus in cells that had experienced an arrest . Due to technical difficulty in amplifying the repeat from a single arrested cell , two approaches were taken . First , we isolated swollen cells ( one indicator of arrest , Figure 3B ) , allowed them to divide to form colonies , and then assessed repeat length in the resulting colonies . To obtain a large enough sample size the analysis was done in the rad52Δ CAG-70 background . 70% of these swollen cells did not form a colony , being permanently arrested or dead . Among the remaining 30% that formed normal-size colonies , there were 14% contractions and 4 . 8% expansions ( Table 2 ) , frequencies that are almost identical to CAG-70 tract instability in rad52Δ colonies not selected for originating from swollen precursor cells ( 15% and 5 . 9% , [12] ) . In the second approach , cells were not preselected , but repeat length was determined only for colonies that grew poorly , which were presumably enriched for cells experiencing arrests . Strikingly , the poorly growing colonies had a large increase in the frequency of total instability , to 67% , compared to 21% for rad52Δ CAG-70 normally growing colonies ( Table 2 ) . The increase in instability was biased toward expansions , with a 2-fold increase in contractions and a significant 6-fold increase in expansions ( Table 2 ) . These results indicate that the cell cycle delays observed in CAG repeat-containing cells are frequently associated with mis-repair events resulting in repeat instability . To determine whether impaired replication fork movement could contribute to repeat instability , we determined CAG-70 tract stability in wild-type cells exposed to a sub-lethal dose of hydroxyurea ( 0 . 1 M ) . Treatment with hydroxyurea resulted in a 7-fold increase in expansions ( to 5 . 6% ) , relative to the untreated control ( 0 . 8% , p<0 . 01 , Table 2 ) . Contractions remained similar to the untreated wild-type control . We conclude that these HU-dependent expansions , which occurred independently of Rad52-HR ( data not shown ) , result from impaired replication across the repeats . These data suggest that the expansions that arose in the rad52Δ poorly growing colonies could be due to slippage events at slowed or restarting replication forks . We showed previously that Rad52-and Dnl4-dependent DSB mis-repair events are other mechanisms for generation of expansions [7] , [12] . The above results show that cell cycle checkpoints are activated in response to damage or interference with fork progression at an expanded CAG repeat . If the level of repeat-induced damage is sufficient or present in a large-enough proportion of cells , phosphorylation of the Rad53 checkpoint kinase should occur . Therefore , we directly tested the phosphorylation status of Rad53 in CAG repeat-containing cells . The wild-type strain harboring a CAG-155 repeat failed to show a detectable shift in mobility of the Rad53 protein , similar to the CAG-0 control ( compare lanes 2 and 3 , Figure 5A ) . Because CAG fragility is proportional to increasing repeat length [29] , [38] , CAG-195 and CAG-240 repeat lengths were also tested , however a phosphorylated Rad53 species was still not detectable ( Figure 5A ) . In contrast , control samples treated with 0 . 05% MMS or 0 . 1 M HU exhibited Rad53 phosphorylation , although the intensity and size of the shifted species was considerably less with HU relative to MMS treatment . Since the wild-type strain has functional repair pathways , the lack of detectable Rad53 phosphorylation could be due to either successful repair of CAG-associated damage or a level of phosphorylated species too low to be visible as a mobility shift on the gel . Indeed , only 3 . 6% of WT CAG-155 cells exhibited arrests >8 hrs ( Figure 2C ) . Therefore , we monitored the Rad53 phosphorylation status in rad52Δ cells that showed a greater frequency of long arrests ( 7 . 8% and 23% for CAG-70 and -155 respectively ) , and elevated CAG fragility [12] . Indeed , rad52Δ CAG-70 or CAG-155 strains but not the rad52Δ CAG-0 control showed a discernible Rad53 phosphorylation response ( Figure 5B ) , indicating that the DNA damage checkpoint is activated in a repeat-dependent manner in this mutant background . The level of Rad53 phosphorylation was repeat-length dependent with the longer CAG-155 repeat eliciting a more robust checkpoint response than the intermediate CAG-70 repeat length as determined by densitometric quantification ( Figure 5B , right ) . We conclude that in the absence of Rad52-dependent repair , the types of damage that occur at an expanded CAG repeat induce a signaling cascade that results in Rad53 phosphorylation and associated downstream checkpoint events . Based on the data from the microcolony and pedigree experiments , it is likely that the same events happen in a wild-type background , but the damage is at a lower level or repaired more quickly so that phosphorylated Rad53 does not accumulate to a level detectable by Western blotting . To determine whether CAG repeat damage was capable of eliciting a cell cycle checkpoint response in the absence of the MRX complex , we determined the Rad53 phosphorylation status in mre11Δ cells . The results revealed two surprising observations . First , Rad53 phosphorylation was observed in mre11Δ CAG-70 and CAG-155 strains ( Figure 5C ) , even though CAG-70 cells , and to a lesser degree CAG-155 cells , appeared to escape cell cycle arrests in this background ( Figure 1 , Figure 2 , Figure 3 ) . Second , Rad53 phosphorylation was also observed in mre11Δ CAG-0 cells , a pattern unlike the WT and rad52Δ CAG-0 controls . Based on these results , we infer that the local checkpoint signaling in response to CAG damage is compromised in mre11Δ cells , while global checkpoint signaling in response to non-repeat damage in the genome is intact . A local Mre11-dependent response at the repeat is further supported by physical detection of the Mre11 protein at the repeat tract , which is enhanced 12-fold compared to a non repeat reference locus by chromatin immunoprecipitation ( ChIP ) ( Figure 5C , right ) . Intriguingly , Mre11 localization to the repeat peaks in S phase , suggesting that the relevant structure sensed by Mre11 is formed during DNA replication . In conclusion , an expanded CAG repeat at a single genomic locus can induce a myriad of cellular arrest responses that depend on signaling via the MRX complex , and culminating in a detectable Rad53 phosphorylation response if the damage is not promptly or efficiently repaired by Rad52-dependent recombination .
Despite the knowledge that expanded CAG repeats interfere with replication , nick ligation , and are fragile sites , direct evidence on whether such damage is at a level or type sufficient to activate DNA damage checkpoints was lacking . This question is especially important since repeated or long checkpoint arrests can affect cell growth potential and lead to apoptosis in higher eukaryotes . It is also of interest to better understand the cellular response to structure-forming sequences , since there are many examples of these sequences in the human genome . Using yeast containing an expanded CAG repeat , we were able to follow the growth potential and fate of single live cells . The presence of a CAG-70 or CAG-155 repeat did not elicit visible Rad53 phosphorylation by Western blot in wild-type cells . We were nonetheless able to detect significant differences between cells with and without an expanded repeat in growth potential , number of arrests , duration of arrests , and morphological abnormalities . Thus even in a wild-type cell , the presence of a long repeat tract was a significant burden on the cell that resulted in measurable effects on growth and division potential . Notably , the arrests that occurred within repeat-containing cells were sometimes of a very long duration , greater than 8 hours and frequently accompanied by severe cell swelling , indicating that a type of damage had occurred that was particularly difficult to repair [21] , [32] , [34] . What is the origin of the damage inducing the checkpoint response ? Recently , it was reported that convergent transcription through CAG repeats induces apoptosis in both dividing and non-dividing human cells [39] . Our CAG repeat is not within a gene or known transcriptional unit , however RT-PCR experiments did show low but equivalent levels of transcript in WT and rad52Δ cells , which could reflect read-through transcription from the neighboring URA3 gene ( M . Koch , J . Yang , and C . H . Freudenreich , data not shown ) . Therefore , it is possible that some of the damage may initiate during transcription . However , the similar transcript levels in the two strains , together with the S-phase delays , the S-phase binding by Mre11p , and the importance of Rad52-dependent repair are all more consistent with the primary damage sensed by the checkpoint occurring during DNA replication . Whatever the initiating event , our data indicate that Rad52-dependent recombination is a key mechanism for overcoming repeat-dependent damage in cells , and without it a strong checkpoint response is induced and cellular growth is severely compromised . For the CAG-155 repeat-containing cells lacking Rad52 , a quarter of the divisions displayed arrests of greater than 8 hours and 36% had morphological abnormalities . In addition , most of those cells had a recurring arrest in the next cell division , indicating that the damage had persisted . The checkpoint responses were all highly repeat-length dependent in the rad52Δ background being significantly greater at CAG-155 compared to CAG-70 , indicating that damage at the longer repeat requires rescue by recombination mechanisms more frequently . In addition , the cell cycle analysis indicated that cells with a CAG-155 tract had a greater tendency to show an S phase delay compared to CAG-70 , a difference exacerbated by Rad52 deficiency . We conclude from this data that the longer repeat has a greater effect on replication , and that a Rad52-dependent process is likely involved in fork restart events for this tract length . This interpretation is supported by comparatively increased CAG-155 fragility observed in a rad52Δ strain [12] . The CAG-70 repeat likely also interferes with replication , as there was a slight increase in S-phase delays , and we showed previously that Srs2-dependent fork reversal occurs at a CTG-55 repeat [7] . Perhaps at this repeat length , fork restart can usually occur without recombination , consistent with genetic results . However if fork reversal or integrity is compromised , recombination may become the preferred pathway , since the CAG-70 expansions that occur in srs2Δ or mre11Δ mutants are Rad52-dependent [7] , [12] . The Dnl4 ligase , needed to complete end-joining repair of DSBs , also played a role in preventing repeat-mediated cell cycle arrests and promoting normal growth and division potential . The requirement for end-joining was more subtle than that for HR , although deficiencies in either process led to a large percentage of cells arrested in G2 with 2N DNA content . While the Dnl4 protein strictly localizes to DSBs , the Mre11 and Rad52 proteins have also been found at stressed or collapsed replication forks and may aid in fork restart [40]-[41] , potentially explaining the greater requirement for these proteins . Altogether , our data are consistent with the idea that the expanded CAG repeat causes multiple types of damage sensed by the checkpoint , including stalled or reversed forks in S phase needing Rad52-dependent restart , and DSBs in G2 that can be repaired by either Rad52-dependent HR or Dnl4-mediated end joining . Importantly , cells which showed an initial arrest response but were able to continue dividing for a limited time to form small , poorly growing colonies showed a significantly elevated frequency of repeat instability . Thus repeat instability may preferentially occur during inefficient or initially failed repair or fork restart events . Cell cycle arrest responses were dramatically altered in the absence of a functional MRX complex . In general , all mre11Δ cells , with or without a repeat tract , are quite compromised for growth with small microcolonies , a reduced division potential , a very high frequency of divisions arresting for >3 hrs ( 90% for CAG-0 ) , and a third arresting for >8 hrs . Notably , although mre11Δ cells were often swollen due to the frequent and long arrests , they did not exhibit many morphological defects . Opposite to the situation in wild-type cells or other DSB repair mutants , the arrest and growth phenotypes were usually less severe in the mre11Δ repeat-containing cells , especially for the CAG-70 tract . Cells containing a CAG-70 tract and lacking the Mre11 protein had a relief of the microcolony growth defect , underwent significantly more divisions , fewer arrests and less cell swelling compared to CAG-0 cells . These results indicate that an intact MRX complex is required for efficient induction of the repeat-mediated checkpoint , and that the majority of CAG-70 damage and some of the CAG-155 damage likely escapes detection by the cell cycle checkpoint machinery in the absence of the MRX complex . This conclusion is further substantiated by the physical detection by ChIP of the Mre11 protein at the repeat tract . Therefore the MRX complex is likely acting as a sensor of damage at the repeat tract , interfacing with a signaling kinase such as Tel1 or Mec1 . Our previous observation of increased CAG fragility in a mec1Δ sml1Δ strain of a similar magnitude to that observed in mre11Δ , whereas a tel1Δ did not increase CAG fragility , suggests that Mec1 is a good candidate for signaling from Mre11 bound damage [12] , [27] . An alternative interpretation is that MRX is needed to create the checkpoint signal , for example by exposing ssDNA that can be coated by RPA . However , our previous results showed that mutation of the Mre11 nuclease activity or associated Exo1 or Sae2 nucleases did not fully recapitulate the mre11Δ phenotype [12] , suggesting that MRX has a function in addition to processing . Interestingly , the overall checkpoint response is not compromised in mre11Δ cells , as constitutive Rad53 phosphorylation was detected , and the CAG-0 arrest phenotypes also indicate a robust and intact global checkpoint response . The S-phase localization of Mre11p to the repeat and the reduced recovery of S phase arrests in the mre11Δ CAG-155 strain ( Figure S4 ) suggest that the primary repeat-induced damage sensed by the MRX complex may arise during replication . Since Mre11p has also been found at HU stalled forks [40] , it may be recognizing the double-strand end at either a reversed or broken fork . The consequences of the absence of MRX sensing are a large increase in CAG fragility , expansions , and cytotoxicity [12] . An interesting and unexpected finding was that the two repeat lengths , CAG-70 and CAG-155 , did not behave identically , suggesting that there are some differences in the DNA structures eliciting the checkpoint at each repeat . Based on the relief of growth and arrest defects by deletion of MRE11 , Mre11 appears to be the primary sensor of damage at CAG-70 . In contrast , cells with a CAG-155 tract still showed some arrest phenotypes in mre11Δ cells , but were particularly dependent on Rad52 for normal growth . Based on these results and previous data that a strain mutated for Mrc1 checkpoint function had a high rate of CAG-155 fragility , we speculate that the longer repeat is detected more efficiently by sensors of fork stalling , such as Mrc1 , making it less dependent on signaling through the MRX complex . It may be that both repeats elicit fork reversal and occasional DSBs that are sensed by MRX , but that the CAG-155 repeat is also able to stall a replication fork long enough to elicit an Mrc1-dependent checkpoint signal . Intriguingly , despite a robust checkpoint response in CAG-155 cells as measured by Rad53 phosphorylation , they formed a slightly larger average microcolony size than cells with a CAG-70 tract . This could be either due to the greater tendency of the CAG-155 strain to adapt ( Table 1 ) , or due to the greater amount of cell swelling ( Figure 3B ) , taking up more space in the microcolony . Perhaps the hypothesized better S-phase structure detection allows for a timelier repair process at the longer repeat . Do other structure-forming sequences elicit similar checkpoint responses ? Expanded CGG/CCG repeats , inverted repeats , and alternative DNA structures such as H-DNA and Z-DNA are hotspots of replication stalling , chromosome breakage and rearrangements , and thus might elicit a similar response [42] . Yet genetic data suggest that expanded CGG repeats may be less efficient at eliciting a checkpoint than CAG repeats , as fork arrest at a CGG repeat was not dependent on the checkpoint function of Mrc1 whereas suppression of CAG fragility and instability is [26] , [28] , [43] . On the other hand , mice with an expanded CGG repeat at the fragile X locus and heterozygous for ATR or ATM exhibit increased frequencies of repeat expansion during intergenerational transmission and in somatic cells [44]-[45] , suggesting that there is some level of checkpoint response to expanded CGG repeats . Variables that could affect the response to different sequences include the nature of the initial damage , processing of the damage , the amount of exposed ssDNA , or the chromatin structure at the repeat . Paradoxically , the ability of CAG repeats to be efficiently recognized by the checkpoint machinery may be helpful in preventing some level of CAG fragility , which is recovered at a lower rate than fragility at expanded CGG/CCG repeats in yeast [46] and has not yet been detected at human disease loci . It will be informative to directly compare the checkpoint responses to CAG versus CGG repeats and other structure-forming sequences in the future . What relevance might our results have for human repeat expansion diseases ? It is known that the RNA and protein products of transcribed and translated CAG/CTG repeats can be toxic to cells . Now we provide the additional knowledge that the expanded DNA itself can be toxic through mechanisms involving DNA replication and DNA damage repair . Since sense or antisense transcription across repeats could contribute to structure formation [47] , the baseline of repeat-induced cytotoxicity may be higher in instances where the CAG repeat locus is also heavily or convergently transcribed [39] . In a multi-cellular organism , many of the long and recurring arrests we observed would probably lead to apoptosis and cell death , the main cause of disease symptoms and morbidity . Indeed , checkpoint activation and cell cycle re-entry have been observed during apoptosis of aging brains of patients with Huntington's as well as other neurodegenerative diseases [48] . A second finding with potential relevance for repeat expansion diseases is that repeat expansions are more frequent in cells undergoing a checkpoint response . Intriguingly , re-entry into the cell cycle after DNA damage can facilitate repair in postmitotic neurons [49] , which could possibly contribute to further repeat expansions . Therefore , DNA repair occurring in the context of an activated checkpoint response may be a cause of repeat expansions in mammalian cells as well .
YAC CF1 harboring CAG repeats is described in [12] , [29] . In this YAC , the CAG repeat is oriented such that the CAG strand is the lagging strand template ( the more stable and expansion-prone orientation ) . For all experiments , yeast strains harboring YAC-CF1 with CAG repeats ( CAG- 70 , 155 , or 195 repeats; Table S2 ) were plated onto YC-Ura-Leu solid media for single colonies and grown for 3 days at 30 °C . CAG repeat length from a portion of the colony was determined by colony PCR [29] . Starting colonies with intact tract lengths were chosen for experiments . 20 µl of overnight culture ( ∼7 . 0 doublings ) was spread as a stripe onto yeast complete solid media lacking Leucine ( YC-Leu ) . Single unbudded , normal-sized G1 cells from the stripe were micromanipulated away to designated locations on the plate using a Nikon Eclipse E400 tetrad dissection scope . Precursor cells were allowed to divide for either 30 hours ( microcolony experiment ) or 18 hours ( pedigree experiment ) at 30 °C . For the microcolony experiment , the growth of precursor cells into microcolonies ( small colonies ) was recorded at 5-hour , 10-hour and 30-hour time intervals . An average of 40 cells ( range 22–73 ) from two experiments were analyzed per strain . Pictures were taken at 10X magnification using an Olympus microscope , and microcolony area at 30 hours was measured using the National Institutes of Health ( NIH ) ImageJ software . Survivors ( plotted ) were defined as area≥0 . 01 mm2; nonsurvivors ( not shown ) were defined as area≤0 . 005 mm2 ( cutoff values in [12] reported incorrectly and corrected here ) . The data set was subjected to non-linear regression analysis and graphed using the Prizm curve-fitting software ( GraphPad Software , San Diego , CA ) . Analysis of variance ( ANOVA ) was used to compare microcolony areas . Fisher's LSD post-hoc test was used to quantify growth differences among CAG -0 , -70 and -155 repeat-containing strains within each genotype . For single cell pedigree analysis , individual divisions within and across pedigrees were monitored for a duration of ∼18 hours ( 5–7 divisions ) on YC-Leu media at 30 °C . This number of divisions was chosen to minimize any confounding effects of senescence and allow a fair comparison between WT and DSB repair mutants that have a compromised division potential ( see Text S1 for further information ) . 15–41 lineages were monitored in parallel on five plates for a total of 31–183 cell divisions; see Table S1 for raw numbers . Mother cells were followed as they do not have a size-related growth delay; daughter cells were discarded . As a rule , precursor cells that failed to initiate growth ( increase in cell volume ) or initiate division were excluded from the experiment since they could have been damaged during micromanipulation . The duration of individual cell division , i . e . growth from a single unbudded cell until separation into daughter cells was recorded; the duration of a normal yeast cell division was set at ≤3 . 0 hours to factor in delays in division introduced by mechanical stress due to micromanipulation , observed to be 30–45 minutes . A bud to mother ratio of <33% was deemed an S phase cell ( small-medium bud ) , or >33% a G2/M phase cell ( large budded ) [50] . Cell cycle position at the G1 , S or G2/M phases was recorded at least twice within a division cycle . Elongated buds were typed as S phase arrests since they occurred after S phase onset . Multibuds were classified as G2/M arrests since they arose after G2 phase onset . Cells that were small budded when a cell cycle delay occurred were categorized as S phase “arrests” , although the S phase checkpoint does not result in a true arrest , but a slowing of S phase and eventual entry into G2 . Cell sizes greater than the size of a normal yeast cell ( >8 µM diameter ) were counted as swollen . Because of interdivisional variation in cell size , the G1 cell of each division was set as the normal size standard for that division , allowing for unbiased assessment of cell swelling among cell divisions . Pictures of swollen cells were taken using a Nikon D40 camera under 80X magnification . Cell area was measured by NIH ImageJ software . Results were graphed using MATLAB version 7 . 9 ( R2009b ) software ( The Mathworks , Natick , MA ) . CAG-70 tracts were chosen to analyze , as expansions are more reliably detected for this length . For the HU experiment , overnight WT CAG-70 cultures were grown for ∼7 . 0 doublings in YC-Leu liquid media +/− 0 . 1 M HU , plated on YC-Leu solid media +/− 0 . 1 M HU , and allowed to form daughter colonies at 30 °C for 3 days . Alternatively , single cells from an overnight YC-Leu culture were micromanipulated on to YC-Leu solid media , and grown for ∼3 days until they attained maximal sizes . A partial normal-sized colony or entire poorly growing colony ( defined as growing to one-third or less the size of a normal colony ) was used for colony PCR using conditions described in [12] . In a subset of colonies showing partial instability , PCR amplification products >10% or >30% of the intensity of the intact band respectively , were counted as an expansion or contraction event . Yeast strains were grown in 1 ml of YC-Leu liquid media with 2% glucose until late-log to early stationary phase . 1 ml of the culture ( ∼1×107 cells ) was pelleted , washed 3X with sterile water , resuspended in cold 70% ( w/v ) ethanol followed by overnight incubation at 4 °C . Cells were subsequently pelleted , resuspended in 50 mM Tris·HCl ( pH 7 . 5 ) buffer containing 1 mg/ml RNaseA , followed by overnight incubation at 37 °C . FACS analysis samples were prepared by pretreatment with 55 mM HCl with 5 mg/ml pepsin , washed and resuspended in FACS buffer ( 200 mM Tris pH 7 . 5; 211 mM NaCl; 78 mM MgCl2 adjusted to pH 7 . 5 with HCl ) containing 1 mg/ml propidium iodide , incubated at −20 °C for 1 hour , transferred to 1 ml of 50 mM Tris pH 7 . 5 and subjected to sonication . The total cellular DNA content from an average of 100 , 000 cells was measured using a FACS-Calibur flow cytometer and BD CellQuest software . FACS plots were generated using ModFit LT software . Total cellular protein was prepared using the trichloroacetic acid ( TCA ) method described in [51] for immunoblotting . Briefly , ∼108 exponentially growing cells ( as determined by OD600 and hemocytometer counting ) were pelleted , washed and resuspended in 20% TCA . Samples were vortexed with glass beads , pelleted at 3000 rpm , boiled in Laemmli buffer ( BioRad ) and the resulting extracts clarified by centrifugation at 3000 rpm . 15 µg of total protein ( quantified by Bradford method ) was loaded per lane; proteins were resolved on an 8% SDS-polyacrylamide gel and transferred to a polyvinylidene difluoride membrane ( GE Amersham ) . The membrane was blocked in 5% milk in TBS-T ( Tween , 0 . 1% ) , incubated with polyclonal , Rad53 primary antibody ( Santa Cruz Biotechnologies ) followed by washes in TBS-T ( 0 . 1% ) and incubation with secondary Anti-goat HRP antibody ( Santa Cruz Biotechnologies ) . Phosphorylated isoforms of Rad53 were visualized by chemiluminescence ( Millipore ) . Semi-quantitative densitometry of phosphorylated Rad53 isoforms was performed using the NIH ImageJ software on film exposures where the signal fell within the linear range . Resultant values were graphed using MS Excel software . A strain with TAP-tagged Mre11 [52] and a CAG-155 repeat was used for ChIP . Yeast cultures grown to an OD600 of 0 . 6 were either unsyncronized ( Asynch ) or synchronized in G1 with α-factor , released into S phase , and samples collected at the indicated time points . Chromatin samples were cross-linked using formaldehyde and processed according to [53] . Mre11-TAP:DNA complexes were immunoprecipitated using rabbit IgG agarose beads directed against protein A on the TAP tag ( Sigma ) [52] . Real-time quantitative PCR was used to amplify a 150 bp fragment 186 bp proximal to the CAG repeat using CAGfor and CAGrev primers in IP and whole cell extract ( WCE ) fractions and an untagged control strain . Similarly , a non-repeat reference locus ( ACT1 ) was amplified from IP , WCE and untagged control fractions ( amplicon length 146 bp , amplified using Act1for2 and Act1rev2 primers; all primers available upon request ) . 2¯-ΔΔct value , i . e . fold enrichment of Mre11-TAP at CAG repeat-containing DNA fragment was obtained by normalizing CAG locus amplification to ACT1 locus amplification in IP and WCE samples . The untagged control showed no enrichment at the CAG locus over the ACT1 locus ( 2¯-ΔΔct value of 1-fold ) . PCR was performed using SYBR-green PCR mix ( BioRad ) on an ABI Prism 7300 sequence detection system . Each PCR reaction was set up in triplicate; PCR cycling conditions- 95 °C for 3 min , 40 cycles of 95 °C for 15 s and 60 °C for 30 s . Asynch , 10 min and 20 min time points represent the average of 3 , 3 , and 4 independent experiments respectively; 0 and 40 min values derived from one experiment . Analysis of variance ( ANOVA ) with a three-way Fisher's LSD post-hoc test was used to compare microcolony areas . Doubling efficiency curves were analyzed using Wilcoxon's sum-rank test . Logistic regression analysis using Chi-square statistics was used to perform 3X2 comparisons on pedigree data sets to determine repeat-specific effects within genotypes; statistically significant ( p<0 . 05 ) interactions were further subjected to a Wald's post-hoc test ( see Table S1 ) . A 2X2 Fisher's exact test was used to compare among genotypes to determine gene specific effects at each CAG repeat length . Analyses were performed using either SAS version 9 . 1 ( SAS Inc . 2003 ) or SPSS Statistics GradPack Version 17 . 0 , 2008 software . | Expansion of a CAG/CTG trinucleotide repeat is the causative mutation for multiple neurodegenerative diseases , including Huntington's disease , myotonic dystrophy , and multiple types of spinocerebellar ataxias . Two reasons for the cell death that occurs in these diseases are toxicity of the repeat-containing RNA and of the polyglutamine-containing protein product . Although the expanded repeat can interfere with DNA replication and repair , it was not known whether the presence of the repeat within the DNA causes any additional cellular toxicity . In this study , we show that an expanded CAG/CTG tract placed within the chromosome of the model eukaryote , budding yeast , elicits a cellular response that interferes with cell growth and division . The effect is enhanced when DNA repair pathways , particularly double-strand break repair , are compromised . Moreover , cells experiencing an arrest were more likely to have undergone further repeat expansions . We show that the conserved MRX protein complex locates to the expanded repeat and is required to sense the damage and activate the DNA damage response . Our results suggest that DNA damage at expanded CAG/CTG repeats could contribute to both tissue degeneration and further repeat instability in affected individuals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neurological",
"disorders/neuromuscular",
"diseases",
"molecular",
"biology/dna",
"replication",
"molecular",
"biology/recombination",
"genetics",
"and",
"genomics/disease",
"models",
"molecular",
"biology/dna",
"repair"
] | 2011 | Expanded CAG/CTG Repeat DNA Induces a Checkpoint Response That Impacts Cell Proliferation in Saccharomyces cerevisiae |
Evolutionary adaptation to a constant environment is often accompanied by specialization and a reduction of fitness in other environments . We assayed the ability of the Lenski Escherichia coli populations to grow on a range of carbon sources after 50 , 000 generations of adaptation on glucose . Using direct measurements of growth rates , we demonstrated that declines in performance were much less widespread than suggested by previous results from Biolog assays of cellular respiration . Surprisingly , there were many performance increases on a variety of substrates . In addition to the now famous example of citrate , we observed several other novel gains of function for organic acids that the ancestral strain only marginally utilized . Quantitative growth data also showed that strains with a higher mutation rate exhibited significantly more declines , suggesting that most metabolic erosion was driven by mutation accumulation and not by physiological tradeoffs . These reductions in growth by mutator strains were ameliorated by growth at lower temperature , consistent with the hypothesis that this metabolic erosion is largely caused by destabilizing mutations to the associated enzymes . We further hypothesized that reductions in growth rate would be greatest for substrates used most differently from glucose , and we used flux balance analysis to formulate this question quantitatively . To our surprise , we found no significant relationship between decreases in growth and dissimilarity to glucose metabolism . Taken as a whole , these data suggest that in a single resource environment , specialization does not mainly result as an inevitable consequence of adaptive tradeoffs , but rather due to the gradual accumulation of disabling mutations in unused portions of the genome .
Evolving populations face the fundamental dilemma that there is no single phenotype that is optimal in all environments . When an evolving population occupies the same selective environment for an extended period of time , no advantage is realized by maintaining fitness on resources it no longer encounters . Adaptation to a selective environment can result in correlated responses in alternative environments . Although these can be synergistic improvements , a response that decreases fitness in other environments is known as specialization . This prevents the rise of “Darwinian demons”: single supergenotypes that are optimized across all conditions [1] . It is critical to understand the origin of specialization because it underlies the origin and maintenance of diversity—it is why “the jack of all trades is a master of none” [2] . Specialization can result from either selective or neutral processes . Antagonistic pleiotropy describes when natural selection favors changes that are beneficial in the current environment but reduce function in other environments . Alternatively , specialization may result from mutations that decrease fitness in alternative environments that are neutral in the selective environment . These mutations have the potential to either drift or hitchhike to fixation via “mutation accumulation . ” As neutral mutations accrue in proportion to the mutation rate , the clearest evidence of mutation accumulation can come from excess specialization in mutator lineages , which contain defects in mutational repair that can elevate mutation rates ∼100-fold [3] . In contrast , where specialization is rapid and occurs in parallel across lineages , a pattern commonly seen for adaptation itself , this has been cited as support of selection-driven antagonistic pleiotropy [4] . The experimental evolution of 12 populations of Escherichia coli grown for thousands of generations on a single substrate has been used to distinguish whether selective or neutral processes drive metabolic specialization [4] . The populations were part of the Lenski Long-Term Evolution Experiment ( LTEE ) [5] , in which wild-type E . coli B have been diluted 1∶100 daily and regrown in well-mixed medium containing glucose as the sole usable carbon source . After 20 , 000 generations ( 20k ) , competitive fitness on glucose had increased by ∼70% . However , respiration assays in static 96-well Biolog plates ( Hayward , CA ) [6] suggested dramatic decreases in metabolic performance on alternative substrates . These declines occurred rapidly and in parallel across populations , coincident with the largest gains in fitness . This was suggested to indicate selection-driven antagonistic pleiotropy as the main mechanism of specialization . Furthermore , because there was only a weak , nonsignificant excess of declines by the populations which had become mutators earlier in the experiment [3] , this suggested neutral mutation accumulation played little role , if any , in glucose specialization by 20k . In this study we have readdressed the basis of specialization in the LTEE populations , motivated by our discovery that the growth rates of isolates in well-mixed media are poorly captured by assays of cellular respiration in static , proprietary media . Given this surprising finding , we analyzed the selected ( i . e . , glucose ) and correlated responses of isolates from both 20k and 50 , 000 generations ( 50k ) from four perspectives:
As a first step to readdressing specialization in the LTEE populations , we sought to replicate the Biolog respiration results at 20k presented by Cooper and Lenski [4] , as well as extend this analysis to the populations at 5ok . Despite changes in the Biolog assay itself , since the previous study , we recovered a similar pattern for the panel of substrates ( Figure S1 ) . The validity of Biolog assays as a proxy for strain improvement came into question after finding decreases for the very substrates on which selection occurred ( Figure 1A ) . Despite abundant evidence of improvement from competitive fitness assays [4] and growth rates [8] , respiration on glucose consistently decreased over the course of the experiment . Furthermore , although one lineage in the A-3 population evolved to utilize as a carbon source the citrate included in Davis Minimal ( DM ) medium [7] , it produced a statistically indistinguishable respiration value from the other Cit− isolates at 50k ( Figure S2 ) . Given that the selective substrates with known fitness improvement had decreased cellular respiration , we turned to direct measurements of growth rates across a wide panel of substrates using a robotic growth analysis system [9] , [10] . By measuring growth rate we capture the demographic metric best correlated with competitive fitness in the evolutionary environment [11] . As the LTEE was performed in shaken , fully aerated flasks , these well-mixed 48-well plates were a closer match to the evolutionary environment than unshaken 96-well plates , as unshaken plates commonly exhibit subexponential growth due to oxygen limitation [10] . Although Biolog uses a proprietary minimal media , for the growth rate measurements we used the same DM media as the LTEE experiment . We omitted citrate , however , as this choice allowed us to include the Cit+ A-3 population in our analysis . We chose carbon sources based upon the substrates for which significant , parallel decreases were previously observed via respiration assays [4] , as well as citrate and several sugars on which growth tradeoffs were previously measured after 2 , 000 generations [12] . Comparing growth rates to respiration data , it becomes evident that the latter is not an accurate assay for growth ( Figure 1B ) . There were many cases where respiration occurred without growth , as 156 out of the 702 strain/substrate combinations measured did not permit growth but did have measurable respiration—a known feature of Biolog assays [6] . Even after removing these categorical disagreements , and a smaller number of instances of growth without respiration ( 11 strain/substrate combinations ) , Biolog respiration values were a poor predictor of growth rate [R2 = 0 . 18 , linear regression F test ( 1 , 499 ) = 108 . 8] . Growth rate data across substrates revealed a surprising degree of correlated gains in performance ( Figure 2 ) . Indeed , at 20k , there were actually more correlated increases in rate than decreases ( 165 versus 99 , respectively , p<0 . 0001 for binomial two-sided test with null of random gains and losses ) . By 50k , the picture had reversed , now with more decreases than increases ( 167 versus 119 , p = 0 . 005 , binomial two-sided test ) . In addition to many quantitative improvements of growth rates , there were several examples where isolates acquired the ability to grow on substrates that the ancestral strain could not utilize over the 48 h time-course of the growth experiments . Only one such example was previously known: the aforementioned gain of citrate utilization by the A-3 population [7] . We found that this strain also gained the ability to grow within 48 h on three C4-dicarboxylate tricarboxylic acid cycle intermediates ( succinate , aspartate , and malate ) . Three other 50k isolates from different replicate populations gained the ability to use this same set of three C4 dicarboxylate intermediates , as well as fumarate . We compared mutators to nonmutators to ask whether mutation accumulation contributed to the observed decreases in growth ( Figure 2 ) . At 20k generations , despite increasing in growth rate more than decreasing ( 47 versus 41 cases ) , the mutators were marginally worse , on average , than nonmutators ( p = 0 . 03 , Pearson's chi-squared test comparing proportion of growth rate reductions ) . By 50k there was a stark pattern of mutators declining in catabolic ability compared to nonmutators ( p<0 . 0001 , Pearson's chi-squared test ) . This can be seen in the large block of blue ( decreases in rate ) for five of the mutators . Nonmutators at 50k still increased in growth rate more often than they decreased ( 80 versus 52 cases , p = 0 . 018 binomial two-sided test ) . Because some strains are known to be affected by the absence of citrate even though they cannot use it as a carbon source [13] , we also tested the growth rate of the 50k strains on alternative substrates supplemented with 1 . 7 mM sodium citrate , as in LTEE growth media . Although the reductions in growth rate relative to the ancestor were ameliorated in some cases by the addition of citrate , the mutator strains still suffered significantly more growth rate decreases than the nonmutators ( p = 0 . 003 , Pearson's chi-squared test ) . Given the trend of both increased and decreased growth rate on alternative carbon sources , we assessed the degree of parallelism with which metabolic erosion occurred . We segmented the data by substrate and asked how many evolved strains decreased in growth rate or cellular respiration on each substrate . We took as a null expectation that decreases in metabolic function are equally as likely as increases , and plotted the observed pattern against this null distribution ( Figure S3 ) . In no case do our observations of metabolic decreases closely match the null distribution . As previously , cellular respiration was reduced for nearly all strains on all substrates . The observed average number of strains with reduced respiration on a substrate was 11 . 3 at 20k and 12 . 8 at 50k , 5 . 3 out of 12 and 6 . 3 out of 13 more strains than would be expected given the null distribution ( p<0 . 0001 , binomial two-tailed exact test ) . For growth rate , at 20k on average 5 . 3 strains reduced in growth rate on each substrate , in fact 0 . 7 fewer strains than expected given the null distribution ( p = 0 . 01 ) . However , this average somewhat masks the bimodal pattern seen in the distribution , with some substrates showing nearly no strains reducing growth rate and others nearly all . At 50k , an average of 8 . 3 strains lost function on each substrate , 1 . 8 more than expected ( p<0 . 0001 ) . Clearly there is some parallelism in decreases in growth rate , but it is worth emphasizing that the substrates used in this study were those for which widespread , parallel losses in cellular respiration were previously observed . We asked whether the correlated changes in performance on alternative substrates could be predicted based on the similarity of the catabolic network for growth on that compound compared to that for glucose . There are two rationales that would support this hypothesis . First , there are more loss-of-function mutations available for a nonglucose substrate if it uses many unique enzymes , and we might expect to see metabolic specialization scale with mutational target size under mutation accumulation . These mutations may either simply be unguarded by purifying selection , or perhaps even selectively advantageous to lose . Second , the balance and direction of flux through various pathways will lead to a different optimal allocation of enzymes to balance the needs of catalysis versus expression costs . As such , antagonistic and synergistic pleiotropy suggest that highly overlapping metabolic flux patterns might be expected to suffer fewer declines , or possibly even synergistic gains , relative to a very differently used substrate . A rough approximation of the similarity between different substrates is to simply group them as sugars or “nonsugars” that require gluconeogenesis for anabolism . To frame this more quantitatively , however , we also used genome-scale metabolic models to make predictions about specialization . In order to approximate internal metabolic states , we used flux balance analysis ( FBA ) to generate predicted flux patterns for each compound [14] . This approach generates a vector of internal flux values that describes the relative flow through every reaction in a cell were it to optimize biomass production per substrate molecule . Although selection in batch culture largely acts upon rate , biomass production per unit substrate has been shown to effectively capture the growth of the LTEE ancestor on glucose , and 50k evolved strains deviated only slightly from this pattern [15] . We therefore compared FBA-derived flux vectors using a number of metrics to determine their degree of dissimilarity to the flux vector for glucose ( see Materials and Methods ) . We first tested whether evolved decreases in growth rate scaled with mutational target size . There is no expected behavior under this hypothesis for increases in growth rate , so we limited our analysis to combinations of strains and substrates for which growth rate had decreased . By identifying the reactions necessary for optimal growth on alternative substrates that are not necessary for growth on glucose , and determining the number of coding nucleotides necessary for those reactions , we were able to approximate the number of available mutations that would decrease growth rate on a substrate . Contrary to our hypothesis , we found no significant relationship between mutational target size and reduction in growth rate [p = 0 . 15 , linear regression F test ( 1 , 146 ) = 2 . 1] ( Figure 3B ) . Our hypothesis also suggests that substrates used more similarly to glucose would permit more rapid growth . Starting with the simple categorization of substrates as sugars and nonsugars , we found no correlation between these groupings and changes in growth rate ( Figure 3A ) . Indeed , there were many reductions in growth rate for sugars other than glucose . To frame this hypothesis in a more quantitative way , we compared the Hamming distance between the vector of predicted fluxes for an alternative compound and that of glucose . Contrary to our hypothesis , we found no significant relationship between metabolic similarity to glucose and correlated responses [p = 0 . 26 , linear regression F test ( 1 , 323 ) = 1 . 27] , and any relationship measured was in fact in the opposite direction as predicted ( Figure 3D ) . As a confirmation that Hamming distance between flux vectors for alternative substrates is biologically relevant , we found that it was a significant predictor of some of the variance in the relative growth rate of the ancestor [p = 0 . 0001 , R2 = 0 . 26 , linear regression F test ( 1 , 48 ) = 17 . 2] ( Figure 3C ) . Alternative metrics to Hamming distance performed similarly poorly in predicting patterns of tradeoffs ( Table S4 ) . These data suggest that the similarity of overall flux patterns is a surprisingly poor predictor about which substrates would experience correlated increases or decreases in performance . Linking the mutator-driven metabolic specialization to their vastly elevated mutation rate itself , we hypothesized that their abundance of amino acid substitutions may generate trends indicative of the types of effects they had upon their gene products . The mutator lineages acquired mutations with a rate up to 0 . 06 per generation [16] . For the A-1 lineage , by 40 , 000 generations there were 627 SNPs , 599 of which were in coding regions , and 513 of those were nonsynonymous [17] . This is a tremendous load of amino acid substitutions , which are viewed as likely to be deleterious due to destabilizing proteins [18]–[20] . We therefore hypothesized that , if protein destabilization was a dominant factor affecting growth at 37°C , we could predictably ameliorate these defects by lowering the growth temperature . Nonmutators will have a small number of such mutations , but the ∼100-fold greater rate of such mutations in the mutator genomes should make growth more temperature sensitive than for nonmutators . Growth rate data support the hypothesis that mutators have general temperature-sensitivity . For 50k isolates , growth rate relative to the ancestor at 30°C was higher than at 37°C in 98 cases , compared to only 37 cases where it was reduced relative to the ancestor ( p<0 . 0001 , binomial two-tailed exact test ) ( Figure 4B ) . That is , despite the fact that these strains have adapted for 50 , 000 generations at 37°C , the ratio of their growth rate to that of the ancestor is higher at the foreign 30°C than their native temperature . This general improvement at the lower temperature was not present for nonmutators ( 56 improved relative to the ancestor by moving to 30°C , 57 worse—p = 0 . 99 , binomial two-tailed exact test ) , and the difference between mutators and nonmutators was significant ( p = 0 . 0002 , Pearson's chi-squared test ) . Furthermore , in the cases where evolved 50k isolates completely lost the ability to grow on substrates , when grown at 30°C these losses were ameliorated more than half the time for mutators ( 35 of 61 ) , significantly more than for nonmutators ( 6 out of 23 , p = 0 . 01 , Pearson's chi-squared test ) ( Figure 4C ) . An alternative hypothesis for the elevated temperature sensitivity of mutators is that the phenotype is directly caused by the mutation in mismatch repair rather than the accumulation of destabilizing mutations that it caused . For the 20k isolates , in most cases growth was better relative to the ancestor at the native 37°C than at 30°C ( 119 versus 75 , p = 0 . 002 , binomial two-tailed exact test ) ( Figure 4A ) . There was no significant difference at 20k between this pattern for mutators and nonmutators ( p = 0 . 20 , Pearson's chi-squared test ) , and no significant difference in the number of rescues from complete loss of growth ( p = 0 . 31 , Pearson's chi-squared test ) , ruling out that mutator status itself generates temperature sensitivity .
Growth rates indicated little evidence in support of widespread antagonistic pleiotropy , with more increases than decreases in growth on alternative compounds through 20k , and very few parallel declines . There were individual counterexamples observed , such as the previously characterized universal loss of ribose utilization early in adaptation [21] , and the tendency for reduced or loss of growth on maltose [22]–[24] . As such , it is clear that examples of antagonistic pleiotropy do exist in the data . However , relatively few other substrates showed this pattern at 20k or 50k , despite the fact that these substrates were those where parallel reductions in respiration were observed . Because selection drives antagonistic pleiotropy , it is commonly expected that the early period of rapid adaptation would coincide with the most tradeoffs in alternative environments , and that the frequent parallelism in the targets among early beneficial mutations would drive parallel losses [4] . Given these criteria , the growth data do not support antagonistic pleiotropy as the primary driver of specialization . There are three implicit assumptions about antagonistic pleiotropy , however , that if not met alter the expectations for specialization driven by selection . First , if different beneficial mutations occur across lineages , they will not necessarily lead to the same pleiotropic tradeoffs . As of 20k , out of the 14 genes screened in all of the populations , there were three genes with mutations in all populations and two more in a majority . The other screened genes had mutations in a minority or none of the other populations , suggesting that a variety of different beneficial mutations occurred across lineages [17] . Second , beneficial mutations in the same target may have differing pleiotropic effects in different lineages due to other mutations present . This “epistatic pleiotropy” [25] has been found to be common in multiple model systems [26]–[29] . Third , the early large-effect beneficial mutations may or may not be responsible for greater pleiotropic effects than later , smaller effect mutations . Yeast morphological pleiotropy scaled with fitness , for example , but the correlation explained only 17% of the variation [30] . The first and second scenarios above—distinct mutations or epistatic pleiotropy—would lead to a scenario whereby parallel metabolic declines are no longer necessarily expected from antagonistic pleiotropy . The third scenario—pleiotropy not scaling with selective effect—would mean the temporal dynamics of fitness gain in the selective conditions and the rate of performance losses in alternative environments need not be tied . These caveats underscore our limited ability to make conclusions about the role of antagonistic pleiotropy in the observed metabolic declines . The only sure determinant of whether a correlated change is the result of pleiotropy or neutral mutation is to genetically manipulate the strains to isolate the effect of individual mutations . This suggests that future experiments , for example , test early and parallel mutations previously screened for epistatic effects [31] for pleiotropy . Ultimately , as we discuss below , the key determinant of the role of mutation accumulation is whether metabolic specialization was substantially affected by mutation rate . Rather than a general pattern of metabolic specialization , these data revealed an unexpected extent of correlated improvements in growth on alternative compounds . Why would E . coli maintain or improve performance on substrates that had not been supplied for decades ? There are three general classes of explanations , two of which mirror the processes considered for pleiotropic tradeoffs . The first explanation for the correlated improvements , and undoubtedly the least likely , would be neutral performance gains through mutations that had no selective consequence in glucose: the beneficial analog of mutation accumulation . If this were the case , mutators might have more increases in rate than nonmutators , and more improvements would have occurred by 50k than 20k , which is the opposite of what was observed . The second explanation of the correlated improvements is that the same mutations that were beneficial during growth on glucose may have led to gains in alternative environments—that is , “synergistic pleiotropy . ” There are known examples of this occurring for early mutations in E . coli evolving in these conditions [12] , [32] , and it is a common pattern seen across organisms ( for example [33] ) . These synergistic mutations may be generally beneficial in the laboratory environment of the LTEE and thus unrelated to carbon source metabolism . Indeed , there are examples of both adaptation to generic aspects of a selective environment , such as the trace metal formulation [34] , and removal or down-regulation of costly genes or genome regions [21] , [35] , [36] . Synergistic pleiotropy could also result from mutations that directly improved glucose metabolism , such as mutations in the phosphotransferase-mediated uptake system that also increased growth on the other sugars imported by this system [12] . A third hypothesis for correlated gains of function on alternative compounds is that there were additional compounds besides glucose ( and citrate ) available from cell excretions or lysis . The serial transfer regime of the LTEE creates a scenario whereby populations use all of their glucose resources within the first few hours , and remain in stationary phase the remainder of the day . The ancestral E . coli excrete a small amount of acetate in this environment , and this increased on average 2-fold by 50k generations [15] . It is thus unsurprising that the strongest , most universal gain in alternative compounds by 20k was on acetate ( Figure 2A ) . In terms of cell lysis , this has allowed one population ( A-2 ) to maintain a long-term polymorphism for over 40 , 000 generations . A “large” colony lineage that grows fast on glucose but lyses substantially in stationary phase cross-feeds a “small” colony lineage that is not as fast on glucose as the larges but has specialized as a “cannibal” [37]–[39] . This results in a stable , negative frequency-dependent fitness effect between these strategies . Although an earlier study of the other populations at 20k failed to reject that fitness interactions were transitive through time [40] , these competitions were performed at a 50∶50 ratio and thus may have missed interactions that occur when one partner is rare . The most remarkable correlated increases were the several examples of “novel” gains of function by evolved isolates on substrates where the ancestor failed to grow . The citrate example has been reported previously [7] , but we did not expect to find other such substrates . These novel gains of function are distinct from what was seen for citrate , as over a longer duration ( >100 h ) the ancestral E . coli seem to grow to measurable density on these substrates . We are currently exploring whether these long lags represent slow physiological acclimation or the emergence of evolved genomic changes . In the case of the Cit+ A-3 lineage , succinate is likely excreted during citrate import [41]; thus , selection for its use is perhaps unsurprising . The fact that several other strains experienced similar gains across the same range of C4 dicarboxylic acids , and that this included the cross-feeding “small” phenotype clone from the A-2 population , appears to suggest that these compounds may be excreted , or present during stationary phase from lysed cells . With the availability of whole-genome scale metabolic models for E . coli , we asked whether we could predict the trend of correlated responses by comparing their pattern of use to that of glucose . We proposed a generic , seemingly obvious hypothesis that the more different the metabolism of an alternative substrate was from the metabolism of glucose , the more likely it would be that populations would have decreased ( or lost ) their ability to use it . As described , this logic holds regardless of whether or not selection drove metabolic specialization . In order to quantify the similarity of substrates , we applied FBA to compare the predicted optimal metabolic flux states for each compound . The data , however , did not support our hypothesis: there was almost no relationship between similarity to glucose and correlated response . Recent in silico attempts to predict growth capability on substrates based on metabolic similarities have had some success [42] , suggesting that evolution may be acting here on functions not included in the model . For example , mutations may have occurred in functions related to differential regulation that distinguish these sugars , rather than central metabolic enzymes for which their use is nearly identical . These mutations are known to have occurred in the LTEE , for example in spoT and nadR [35] , [43] . These predictions are also based on the assumption that glucose is the only available carbon source . If growth on other carbon sources is under selection due to their excretion or presence after cell lysis , it may explain some of the lack of predictive power here . Although the identity of the substrates that experienced tradeoffs ( or improvements ) were not those we expected , we reasoned that the biophysical effects of the deluge of mutations in mutators might lead to a predictable pattern of temperature sensitivity in these strains . The genomic sequences and data available to date [16] , [17] suggest mutators will have on the order of 500–2 , 000 nonsynonymous mutations , perhaps more . Random amino acid substitutions have been shown to be mildly deleterious in general due to destabilizing proteins [20] . To ask whether tradeoffs observed in the mutators were at least partly due to destabilizing mutations in proteins needed for alternative substrates , we tested whether mutators would be more sensitive to changes in incubation temperature than nonmutators . Consistent with this hypothesis , we found that the 50k mutators performed better relative to the ancestor at 30°C than at the 37°C temperature where they have evolved . One alternative hypothesis that was ruled out is that the mutator allele itself leads to temperature sensitivity , as the 20k mutators performed better at 37°C than at 30°C , and the changes with temperature were not distinguishable from nonmutators . Although other alternative hypotheses may explain some of the temperature sensitivity , these data are consistent with the hypothesis that neutral degradation of protein-coding sequences in these strains proceeded via partial destabilization on the way to eventual loss of function . The comparison of mutators and nonmutators at 50k strongly suggests that neutral mutation accumulation was the primary driver of metabolic specialization . The difference in metabolic erosion allows us to distinguish the overall trend from forms of antagonistic pleiotropy that could have lacked parallelism ( different mutations or epistatic pleiotropy ) or that may have arisen late relative to fitness gains ( if pleiotropy did not scale with selective effects ) . Despite a significant difference from nonmutators in the proportion of growth rate reductions , after 20 , 000 generations and a decade of adaptation the mutators still increased growth rate in more cases than they decreased , and only by 50k did mutators as a group have more decreases than increases . By the later time point , five of the seven mutators had decreased growth ( or complete loss ) for essentially every single alternative compound ( except citrate and C4 dicarboxylic acids for A-3 , which were under selection for this strain ) . Interestingly , the other two ( A-1 , A-2S ) do not show this pattern . These counterexamples may be due to the fact that A-1 acquired its mutator status late [17] , and A-2S is the cross-feeding generalist described above that adapted to grow upon lysed cell material [38] . The late appearance of metabolic erosion argues for the unparalleled utility of truly long-term experiments . A neutral process such as mutation accumulation needs time to become apparent , although hitchhiking with beneficial mutations can speed their fixation ( i . e . , “draft” [44] , [45] ) . With a reduced effective population size , the window of selective effects that behave neutrally grows . As such , the effects of elevated mutation rates and mutation accumulation become apparent much more quickly with evolution regimes with small bottlenecks , such as single colonies [46] , [47] . The late appearance of specialization also contrasts sharply with abundant evidence that lineages can diversify and specialize in mere tens or hundreds of generations . In addition to population size , this difference in timescale appears to correlate with the type of selective environment , and thus the evolutionary process that was responsible . Whereas the Lenski LTEE is notable as an environment with a single nutrient resource at high concentration , the cases of rapid diversification have involved spatial heterogeneity [48] , rate-limiting resources in a chemostat [49] , or the presence of multiple substrates simultaneously [50] . In those scenarios , selection is actively pulling on different performance features of an organism and antagonistic pleiotropy appears to dominate . The relatively slow degradation of catabolic capacity in the LTEE suggests that E . coli faces comparatively little tension between improving upon glucose and maintaining performance on other substrates , even those which are predicted to be utilized in a very distinct manner . Specialization in this case appears not to have been a requisite tradeoff of adaptation , but rather a result of the degradation of unneeded proteins . Given the severity of metabolic erosion for mutators even in large laboratory populations , it is remarkable just how common mutator lineages are in nature . Mutators have been isolated at frequencies over 1% and seem to be particularly common in organisms such as pathogens [51] . The frequency of mutators in nature , despite the associated costs , may be partially explained by increased evolvability , shown in laboratory medium [52] and in mice [53] . Our results add substantially to the idea that an elevated mutation rate is an ill-fated long-term strategy even for large populations , as declines in performance in alternative environments will eliminate previously occupied parts of the niche space . Recent findings have suggested that mutators may tend to attenuate their increased mutation rate over time [16] , [54] , [55] , perhaps to avoid the harmful effects of Muller's Ratchet . Thus both tradeoffs in alternative environments and mutation load in selective environments may contribute to the paradox that over the short term lineages often benefit from elevated mutation rates , but the long-term trend across phylogenies has been for stability in mutation rates of free-living microbes [56] .
E . coli B isolates were obtained from the LTEE [5] after 20 , 000 and 50 , 000 generations . Briefly , in the evolution experiment 12 populations of E . coli B were founded with either the arabinose-negative strain REL606 ( populations A-1 to A-6 ) or the otherwise isogenic arabinose-positive derivative , REL607 ( A+1 to A+6 ) . These have been evolved since 1988 in 50 mL flasks containing 10 mL of DM media with 139 µM glucose ( 25 mg/L ) as a carbon source . The cultures were grown at 37°C while shaking at 120 rpm , and were transferred daily via 1∶100 dilutions ( ∼6 . 64 net doublings per day ) . The isolates analyzed in this experiment consisted of the ancestral lines REL606 and REL607 , as well as the “A” clone frozen at 50k and 20k generations for the 12 populations . The A-2A clones at 20k and 50k were from the “large” lineage that has coexisted with a cross-feeding “small” lineage for tens of thousands of generations [57] , and thus here we refer to them as A-2L . At 50k we also examined A-2C ( REL11335 ) , a “small” clone that we refer to here as A-2S . All evolved strains are listed in Table S1 . For growth rate measurements , we acclimated out of the freezer by inoculating 10 µL frozen cultures into 630 µL modified DM250 media in 48-well micotiter plates ( Costar ) and growing overnight on a plate shaking tower ( Caliper ) . All growth rates were measured at the LTEE selective temperature ( 37°C ) unless otherwise described . The modified DM media is the same as previously used throughout the evolution of these strains [5] , except it contained 250 mg/L glucose and no sodium citrate . Following acclimation , saturated cultures were transferred into new plates with a 1∶64 dilution in DM media supplemented with 5 mM of a single carbon source . Under these conditions , growth in plates correlates well with growth in flasks , both with and without citrate [p<0 . 0001 , R2 = 0 . 74 , linear regression F test ( 1 , 28 ) = 79 . 3] . The substrates analyzed were those where consistent reduction in cellular respiration were previously observed , as well as several sugars for which fitness changes had been previously measured after 2 , 000 generations [12] and citrate . Between 3 and 11 biological replicates were run for each strain/carbon source combination . Optical densities were obtained every 30 min to 1 h on a Wallac Victor 2 plate reader ( Perkin-Elmer ) , until 48 h had passed or cultures reached saturation , using a previously described automated measurement system [9] , [10] . Growth rates were determined by fitting an exponential growth model using custom analysis software , Curve Fitter ( N . F . Delaney , CJM , unpublished; http://www . evolvedmicrobe . com/Software . html ) . Representative growth curves and fitted growth rates are shown in Figure S4 . The growth rate for all strains relative to the ancestor was calculated for each plate ( averaging over the two ancestors , REL606 and REL607 ) , and averaged across plates . Mean growth rates relative to the average of the ancestors were used throughout . When quantifying the number of increases and decreases in rate for evolved strains , we used all of the data for the substrates for which the ancestor exhibited growth—a necessary criterion for the evolved strains to demonstrate reductions . Biolog assays for respiration capacity were run as in Cooper and Lenski [4] . Briefly , cultures were grown from freezer stocks for two cycles of 1∶100 dilution and 24 h of growth in 10 mL LB in flasks shaking at 37°C . LB was used to avoid catabolite repression due to growth in minimal media , which could result in fewer positive results on nonglucose carbon sources . These cultures were inoculated 1∶100 into fresh LB and grown for 6 h before being spun down at 12 , 000 g for 10 min and rinsed in saline to remove residual medium . Rinsed cells were resuspended in IF-0 buffer with dye added ( Biolog ) to a constant density of 85% transmittance , and all wells of Biolog PM1 plates were inoculated with 100 µL of this suspension . The plates were incubated , unshaken , at 37°C . OD600 was measured at 0 , 4 , 12 , 24 , and 48 h , and all well readings were adjusted by subtracting the reading of the well at 0 h . A trapezoidal area approximation [4] combined the five measurements for each well into one value , which reflects the area under the curve ( AUC ) of optical density versus time . One replicate plate experiment was performed for each evolved strain at 20k and 50k generations , and four replicates were run for each ancestor ( REL606 and REL607 ) . Tests for tradeoffs for the evolved strains as a group on a substrate had were one-sample t tests against the ancestral distribution with significance cutoff of p = 0 . 002 to adjust for multiple comparisons . Tests for individual strains were the same but with a cutoff of p = 0 . 0005 . Flux analysis was carried out with a genome-scale model of E . coli metabolism ( iAF_1260 [58] ) . The model incorporates 2 , 382 reactions and 1 , 668 metabolites . The default minimal media environment and reaction bounds were used . Fluxes were predicted for each individual carbon source provided , normalized by number of carbon atoms to 10 units of glucose . Maximal biomass per substrate was used as the objective criterion as previously described [15] . To determine whether a reaction was necessary for optimal growth on a substrate , each reaction flux predicted was individually constrained to zero . Only the necessary reactions , those for which constraining the flux resulted in a reduction in final biomass , were considered in the analysis of differences between flux vectors ( Table S2 ) . Reaction differences between substrates , considered for the Hamming distance , are listed in Table S3 . Table S4 summarizes alternative distance metrics that were used to assess the difference between flux vectors . | Adaptation to a single constant environment is commonly expected to result in decreased performance in alternative conditions , or specialization . It has been proposed that , rather than occurring through the neutral accumulation of mutations in unused alternative pathways , this happens because loss of these pathways enhances fitness in the constant environment via “tradeoffs . ” We examined growth rates across a variety of nutrients for 12 independent lineages of Escherichia coli that had evolved in the laboratory for decades in a glucose-containing medium . Surprisingly , after 20 , 000 generations there were actually widespread improvements in the use of alternative nutrients , rather than the expected declines . After 50 , 000 generations , however , we find that this trend reversed for those populations that evolved a much higher mutation rate . This indicates that high mutation rate , and not adaptive tradeoffs per se ( as had been previously proposed ) , is the primary driver of specialization . These results caution against general assumptions about the importance of adaptive tradeoffs during evolution , and emphasize the key role that newly evolved changes in mutation rate can play in promoting niche specialization . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"organismal",
"evolution",
"microbial",
"metabolism",
"microbial",
"mutation",
"population",
"genetics",
"metabolic",
"networks",
"microbiology",
"bacterial",
"biochemistry",
"escherichia",
"coli",
"parallel",
"evolution",
"mutation",
"prokaryotic",
"models",
"model",
"organisms",
"microbial",
"evolution",
"population",
"biology",
"forms",
"of",
"evolution",
"microbial",
"physiology",
"biology",
"genetic",
"drift",
"systems",
"biology",
"biochemistry",
"microevolution",
"bacterial",
"evolution",
"adaptation",
"natural",
"selection",
"computational",
"biology",
"evolutionary",
"biology",
"metabolism",
"evolutionary",
"processes"
] | 2014 | Metabolic Erosion Primarily Through Mutation Accumulation, and Not Tradeoffs, Drives Limited Evolution of Substrate Specificity in Escherichia coli |
In the life sciences , many measurement methods yield only the relative abundances of different components in a sample . With such relative—or compositional—data , differential expression needs careful interpretation , and correlation—a statistical workhorse for analyzing pairwise relationships—is an inappropriate measure of association . Using yeast gene expression data we show how correlation can be misleading and present proportionality as a valid alternative for relative data . We show how the strength of proportionality between two variables can be meaningfully and interpretably described by a new statistic ϕ which can be used instead of correlation as the basis of familiar analyses and visualisation methods , including co-expression networks and clustered heatmaps . While the main aim of this study is to present proportionality as a means to analyse relative data , it also raises intriguing questions about the molecular mechanisms underlying the proportional regulation of a range of yeast genes .
Relative abundance measurements are common in molecular biology: nucleic acids typically have to be provided at a set concentration for sequencing or microarray analysis; sequencing methods report a large but finite total of reads , of which any particular sequence is a proportion . Sometimes , researchers are interested in the relative abundance of different components . Other times , they have to make do with relative abundance to gain insight into the system under study . Whatever the case , data that carry only relative information need special treatment . Awareness is growing [1 , 2 , 3] but it is not yet widely appreciated that common analysis methods—including correlation—can be very misleading for data carrying only relative information . Compositional data analysis [4] ( CoDA ) is a valid alternative that harks back to Pearson’s observation [5] of ‘spurious correlation’ , i . e . , while statistically independent variables X , Y , and Z are not correlated , their ratios X/Z and Y/Z must be , because of their common divisor . ( Note: this differs from the logical fallacy that “correlation implies causation” . ) Proportions , percentages and parts per million are familiar examples of compositional data; the fact that the representation of their components is constrained to sum to a constant ( i . e . , 1 , 100 , 106 ) emphasizes that the data carry only relative information . Note that compositional data do not necessarily have to sum to a constant; what is essential is that only the ratios of the different components are regarded as informative . Correlation—Pearson , Spearman or other—leads to meaningless conclusions if applied to compositional data because its value depends on which components are analyzed [4] . Problems with correlation can also be demonstrated geometrically ( Fig . 1 ) : the bivariate joint distribution of relative abundances says nothing about the distribution of absolute abundances that gave rise to them . Thus , relative data is also problematic for mutual information and other distributional measures of association . To further illustrate how correlation can be misleading we applied it to absolute and relative gene expression data in fission yeast cells deprived of a key nutrient [6] . How then can we make sound inferences from relative data ? We show how proportionality provides a valid alternative to correlation and can be used as the basis of familiar analyses and visualizations . We conclude by putting this analysis strategy in perspective , discussing challenges , caveats and issues for further work , as well as the biological questions raised in this study .
Our results are based on data from Marguerat et al . [6] on the absolute levels of gene expression ( i . e . , mRNA copies per cell ) in fission yeast after cells were deprived of a key nutrient ( Fig . 2 ) . Unlike many experiments where researchers ensure ( or assume ) cells produce similar amounts of mRNA across conditions [7] , this experiment ensured cells produced very different amounts so as to illustrate the merits of absolute quantification ( S1 Fig . ) . Total abundance may vary dramatically in other experimental settings—such as in comparing diseased and normal tissues , tissues at different stages of development , or microbial communities in different environments . To illustrate the key points of this paper , we worked with positive data only ( i . e . , we excluded records with any zero or NA values ) : measurements of 3031 components ( i . e . , mRNAs ) at 16 time points . Furthermore , we applied analysis methods ( specifically , correlation ) to the absolute abundance data without transformation ( e . g . , taking logarithms ) because we believe this approach yields useful insights and simplifies the presentation of the central ideas of this paper ( see [8] and S1 Supporting Information ) . Before looking at issues with pairs of components , it is important to note that interpreting differences in the relative abundance of a single component can be challenging . Tests for differential expression are popular for analyzing relative data in bioscience . Much attention has been given to dealing with small numbers of observations and large numbers of tests , but comparatively little to “…the commonly believed , though rarely stated , assumption that the absolute amount of total mRNA in each cell is similar across different cell types or experimental perturbations” [7] . The relationship between the relative and absolute abundance of a component can be understood in terms of fold change over time . When total absolute abundance of mRNA stays constant , fold changes in both absolute and relative abundance of each mRNA are equal . When total absolute abundance varies , fold changes in absolute and relative abundances of each mRNA are no longer equal and can change in different directions . Between 0 and 3 hours there were 1399 yeast mRNAs whose absolute abundance decreased , and whose relative abundance increased . Clearly , mRNAs are being expressed differently , but to describe them as “under- or over-expressed” is too simplistic—here lies the interpretation challenge ( see S1 Supporting Information ) . While “differential expression” of relative abundances is challenging to interpret , in the absence of any other information or assumptions , correlation of relative abundances is just wrong . We stress in the absence of any other information or assumptions to highlight the common assumption of constant absolute abundance of total mRNA across all experimental conditions . If this assumption holds , and all the mRNAs comprising that total are considered , the relative abundance of each kind of mRNA will be proportional to its absolute abundance , and analyses of correlation or “differential expression” of the relative values will have clear interpretations . The revisitation of this assumption [7] should raise alarm bells about the inferences drawn from many gene expression studies . Fig . 1 ( a ) shows why correlation between relative abundances tells us nothing about the relationship between the absolute abundances that gave rise to them: the perfectly correlated relative abundances could come from any set of absolute abundance pairs that lie on the rays from the origin . This many-to-one mapping means that other measures of statistical association ( e . g . , rank correlations or mutual information ) will not tell us anything either when applied to purely relative data . But is this problem just a theoretical construct ? A rare issue ? Consider the red mRNA pair in Fig . 2: while their absolute abundances over time are strongly positively correlated , if someone ( inappropriately ) used correlation to measure the association between the relative abundances of these two mRNAs they would form the opposite view ( Fig . 3 ( a ) ) ; correlation between the blue mRNA pair in Fig . 2 is similarly misleading ( S2 Fig . ) . What of the other 4 . 5 million pairs of mRNAs ? Fig . 3 ( b ) summarizes all discrepancies between correlations of absolute abundance , and correlations of relative abundance , showing clearly that the apparent correlations of relative abundances tell a very different story from those of the absolute data . So how should we go about analyzing these relative data ? CoDA theory provides three principles [4 , 9]: Scale invariance: analyses must treat vectors with proportional positive components as representing the same composition ( e . g . , ( 2 , 3 , 4 ) is equivalent to ( 20 , 30 , 40 ) ) Subcompositional coherence: inferences about subcompositions ( subsets of components ) should be consistent , regardless of whether the inference is based on the subcomposition or the full composition . Permutation invariance: the conclusions of analyses must not depend on the order of the components . Correlation is not subcompositionally coherent: its value depends on which components are considered in the analysis , e . g . , if you deplete the most abundant RNAs from a sample [10] and use correlation to measure association between relative abundances , you get different correlations to the undepleted sample ( S3 Fig . ) . Proportionality obeys all three principles for analyzing relative data . If relative abundances x and y are proportional across experimental conditions i , their absolute abundances must be in proportion: x i t i ∝ y i t i ⇒ x i ∝ y i where ti is the total abundance in condition i ( Fig . 1 ( b ) ) . We proposed a “goodness-of-fit to proportionality” statistic ϕ to assess the extent to which a pair of random variables ( x , y ) are proportional [11] . ϕ is related to logratio variance [4] , var ( log ( x/y ) ) , and is zero when x and y behave perfectly proportionally . However , when x and y are not proportional , ϕ has both a clear geometric interpretation and a meaningful scale , addressing concerns raised about logratio variance [3]: the closer ϕ is to zero , the stronger the proportionality . We consider “strength” of proportionality ( goodness-of-fit ) rather than testing the hypothesis of proportionality because it allows us to compare relationships between different pairs of mRNAs ( S1 Supporting Information ) . We calculated ϕ for the relative abundances of all pairs of mRNAs and compared it to the correlations between their absolute abundances ( S4 Fig . ) : clearly , the absolute abundances of most mRNA pairs are strongly positively correlated; far fewer are also strongly proportional . Focusing on these strongly proportional mRNAs , we extracted the 424 pairs with ϕ < 0 . 05 . We graphed the network of relationships between these mRNAs ( S5 Fig . ) , an approach similar to gene co-expression network [12] or weighted gene co-expression analysis [13] but founded on proportionality and therefore valid for relative data . The network revealed one cluster of 96 , and many other smaller clusters of mRNAs behaving proportionally across conditions . Using ϕ as a dissimilarity measure , we formed heatmaps of the three largest clusters ( S6 and S7 Figs . ) similar to the method of Eisen et al . [14] but , again , using proportionality not correlation .
Other researchers have recognized the compositional nature of molecular bioscience data , including [15] as discussed in [16] . Strategies have been proposed to ameliorate spurious correlation in the analysis of relative abundances [2 , 3] . We contend that there is no way to salvage a coherent interpretation of correlations from relative abundances without additional information or assumptions; our argument is based on Fig . 1 . ReBoot [2] attempts to establish a null distribution of correlations against which bootstrapped estimates of correlations can be compared . Aitchison articulates problems with this approach [4 , p . 56–58] . SparCC [3] injects additional information by assuming the number of different components is large and the true correlation network is sparse . This equates to assuming “that the average correlations [between absolute abundances] are small , rather than requiring that any particular correlation be small” [3 , Eq . 14] . This means the expected value of the total absolute abundance will be constant ( as the sum of many independently distributed amounts ) . We are concerned with situations where that assumption cannot be made , or where the aim is to describe associations between relative amounts . We are also keen to raise awareness that correlation ( and other statistical methods that assume measurements come from real coordinate space ) should not be applied to relative abundances . This is highly relevant to gene coexpression networks [12] . Correlation is at the heart of methods like Weighted Gene Co-expression Network Analysis [13] and heatmap visualization [14] . These methods are potentially misleading if applied to relative data . This concern extends to methods based on mutual information ( e . g . , relevance networks [17] ) since , as Fig . 1 shows , the bivariate joint distribution of relative abundances ( from which mutual information is estimated ) can be quite different from the bivariate joint distribution of the absolute abundances that gave rise to them . Measures of association produce results regardless of the data they are applied to—it is up to the analyst to ensure that the measures are appropriate to the data . Currently , there are many gene co-expression databases available that provide correlation coefficients for the relative expression levels of different genes , generally from multiple experiments with different experimental conditions ( see e . g . , [18] ) . As far as we are aware , none of the database providers explicitly address whether absolute levels of gene expression were constant across experimental conditions . If the answer to this question is “no” , we would not recommend these correlations be used for the reasons demonstrated in this paper . If the answer is “yes” we still advocate caution in applying correlation to absolute abundances for reasons discussed in S1 Supporting Information . While the main aim of this study is to present and illustrate principles for analyzing relative abundances , it has also uncovered intriguing biological insight with respect to gene regulation . The largest cluster of proportionally regulated mRNAs ( 96 genes , S1 Supporting Information ) was highly enriched for mRNAs down-regulated as part of the core environmental stress response [19] , including 66 mRNAs that encode ribosomal proteins , and the remaining mRNAs also associated with roles in protein translation , such as ribosome biogenesis , rRNA processing , tRNA methyltransferases and translation elongation factors . The absolute levels of these mRNAs decrease after removal of nitrogen [6] . The notable coherence in biological function among the mRNAs in this cluster is higher than typically seen when correlative similarity metrics for clustering are applied ( e . g . , [19] ) . These 96 mRNAs show remarkable proportionality to each other over the entire timecourse ( S8 Fig . ) , and maintain near constant ratios across all conditions ( S9 Fig . ) . Given the huge energy invested by yeast cells for protein translation ( most notably ribosome biogenesis [20 , 21] , it certainly makes sense for cells to synchronize the expression of relevant genes such that translation is finely tuned to nutritional conditions . Evidently , numerous ribosomal proteins and RNAs function together in the ribosome , demanding their coordinated expression; more surprisingly , multiple other genes , with diverse functions in translation , show equally pronounced proportional regulation across the timecourse . These findings raise intriguing questions as to the molecular mechanisms underlying this proportional regulation , suggesting sophisticated , coordinated control of numerous mRNAs at both transcriptional and post-transcriptional levels of gene expression . While proportionality and the ϕ-statistic provide a valid alternative to correlation for relative data , there are still some challenges in their application . First is the treatment of zeroes , for which there is currently no simple general remedy [22] . Second , and related , is the fact that “many things that we measure and treat as if they are continuous are really discrete count data , even if only at the molecular extremes” [23] and count data is not purely relative—the count pair ( 1 , 2 ) carries different information than counts of ( 1000 , 2000 ) even though the relative amounts of the two components are the same . Correspondence analysis [24] , or methods based on count distributions ( e . g . , logistic regression and other generalized linear models ) may provide ways forwards .
All data and code [25] needed to reproduce the analyses and visualizations set out in this paper are contained in the Supporting Information , along with additional illustrations and detailed explanations . The “goodness-of-fit to proportionality” statistic ϕ can be used to assess the extent to which a pair of random variables ( x , y ) are proportional [11] . Aitchison [4] proposed logratio variance , var ( log ( x/y ) ) , as a measure of association for variables that carry only relative information . When x and y are exactly proportional var ( log ( x/y ) ) = 0 , but when x and y are not exactly proportional , “it is hard to interpret as it lacks a scale . That is , it is unclear what constitutes a large or small value… ( does a value of 0 . 1 indicate strong dependence , weak dependence , or no dependence ? ) ” [3] . Logratio variance can be factored into two more interpretable terms: var ( log ( x / y ) ) = var ( log x − log y ) = var ( log x ) + var ( log y ) − 2 cov ( log x , log y ) ( 1 ) = var ( log x ) ⋅ ( 1 + var ( log y ) var ( log x ) − 2 var ( log y ) var ( log x ) cov ( log x , log y ) var ( log x ) var ( log y ) ) = var ( log x ) ⋅ ( 1 + β 2 − 2 β | r | ) ≜ var ( log x ) ⋅ ϕ ( log x , log y ) ( 2 ) where β is the standardized major axis estimate [26] of slope of random variables log y on log x , and r the correlation between those variables . The first term in Equation 2 , var ( log x ) , is solely about the magnitude of variation at play and has nothing to do with y . The second term , ϕ , describes the degree of proportionality between x and y , and forms the basis of our analysis of the relationships between relative values . Other non-negative functions of β and r that are zero when x and y are perfectly proportional could be formed; this is described in more detail in S1 Supporting Information , as well as why ϕ is preferable to an hypothesis testing approach . There is no need to calculate β or r to assess strength of proportionality; they simply provide a clear geometric interpretation of ϕ; in practice , one can use the relationship ϕ ( log x , log y ) = var ( log ( x/y ) ) /var ( log x ) . The ϕ statistic is a measure of goodness-of-fit to proportionality that combines two quantities of interest: β , the slope of the line best describing the relationship between random variables log x and log y; and r , whose magnitude estimates the strength of the linear relationship between log x and log y . “Goodness-of-fit” describes how well a statistical model fits a set of observations and is a familiar concept in regression , including linear and generalised linear models , but note that ϕ—specifically the slope ( β ) of the standardized major axis—is motivated by allometry rather than regression modeling . We are interested in assessing whether two variables are directly proportional , rather than predicting one from the other: “use of regression would often lead to an incorrect conclusion about whether two variables are isometric or not” [26 , p . 265] . Note also that ordinary least squares regression fits are not symmetric: in general , the slope of y regressed on x is different to the slope of x regressed on y [27] . While goodness-of-fit measures for regression may not generally be appropriate for assessing proportionality , Zheng [28] explores the concordance correlation coefficient ρc [29] which could be modified to provide an alternative measure of proportionality defined as ρ p ( log x , log y ) ≜ 2 cov ( log x , log y ) var ( log x ) + var ( log y ) and related to var ( log ( x/y ) ) by the terms in Equation 1 . This “proportionality correlation coefficient” ranges from −1 ( perfect reciprocality ) to +1 ( perfect proportionality ) and lacks the clear geometric interpretation of ϕ . We have used ϕ ( log x , log y ) to emphasize the relationship between ϕ and logratio variance . However to ensure that the ϕ values for component pair ( i , j ) are on the same scale ( i . e . , comparable to ) the ϕ values for component pair ( m , n ) , it is necessary to use the centered logratio ( clr ) transformation instead of just the logarithm ( S1 Supporting Information ) . The clr representation of composition x = ( x1 , … , xi , … , xD ) is the logarithm of the components after dividing by the geometric mean of x: clr ( x ) = log x 1 g m ( x ) , ⋯ , log x i g m ( x ) , ⋯ , log x D g m ( x ) , ensuring that the sum of the elements of clr ( x ) is zero . Note that dividing all components in a composition by a constant ( i . e . , the geometric mean gm ( x ) ) does not alter the ratios of components . Gene co-expression networks [12 , 13] are generally based on a pairwise distance or dissimilarity matrix which is often a function of correlation and thus not appropriate for relative data . Proportionality is appropriate , but ϕ does not satisfy the properties of a distance—most obviously , it is not symmetric unless β = 1: ϕ ( log x , log y ) = 1 + β 2 - 2 β | r | ϕ ( log y , log x ) = 1 + 1 β 2 - 2 1 β | r | . We are most interested in pairs of variables where β and r are near 1 and want to preserve the link between ϕ ( log x , log y ) , β and r . Hence , our approach to forming a dissimilarity matrix is simply to work with ϕ ( log xi , log xj ) where i < j , in effect , the lower triangle of the matrix of ϕ values between all pairs of components . This symmetrised form of ϕ was then used to lay out a network of the 145 mRNAs that were involved in 424 pairwise relationships with ϕ < 0 . 05 . We used the symmetrised form of ϕ as the basis of the cluster analysis and heatmap expression pattern display ( e . g . , S10 Fig . ) described by Eisen et al . [14] . | Relative abundance data is common in the life sciences , but appreciation that it needs special analysis and interpretation is scarce . Correlation is popular as a statistical measure of pairwise association but should not be used on data that carry only relative information . Using timecourse yeast gene expression data , we show how correlation of relative abundances can lead to conclusions opposite to those drawn from absolute abundances , and that its value changes when different components are included in the analysis . Once all absolute information has been removed , only a subset of those associations will reliably endure in the remaining relative data , specifically , associations where pairs of values behave proportionally across observations . We propose a new statistic ϕ to describe the strength of proportionality between two variables and demonstrate how it can be straightforwardly used instead of correlation as the basis of familiar analyses and visualization methods . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Proportionality: A Valid Alternative to Correlation for Relative Data |
MicroRNAs ( miRNAs ) and small nucleolar RNAs ( snoRNAs ) are two classes of small non-coding regulatory RNAs , which have been much investigated in recent years . While their respective functions in the cell are distinct , they share interesting genomic similarities , and recent sequencing projects have identified processed forms of snoRNAs that resemble miRNAs . Here , we investigate a possible evolutionary relationship between miRNAs and box H/ACA snoRNAs . A comparison of the genomic locations of reported miRNAs and snoRNAs reveals an overlap of specific members of these classes . To test the hypothesis that some miRNAs might have evolved from snoRNA encoding genomic regions , reported miRNA-encoding regions were scanned for the presence of box H/ACA snoRNA features . Twenty miRNA precursors show significant similarity to H/ACA snoRNAs as predicted by snoGPS . These include molecules predicted to target known ribosomal RNA pseudouridylation sites in vivo for which no guide snoRNA has yet been reported . The predicted folded structures of these twenty H/ACA snoRNA-like miRNA precursors reveal molecules which resemble the structures of known box H/ACA snoRNAs . The genomic regions surrounding these predicted snoRNA-like miRNAs are often similar to regions around snoRNA retroposons , including the presence of transposable elements , target site duplications and poly ( A ) tails . We further show that the precursors of five H/ACA snoRNA-like miRNAs ( miR-151 , miR-605 , mir-664 , miR-215 and miR-140 ) bind to dyskerin , a specific protein component of functional box H/ACA small nucleolar ribonucleoprotein complexes suggesting that these molecules have retained some H/ACA snoRNA functionality . The detection of small RNA molecules that share features of miRNAs and snoRNAs suggest that these classes of RNA may have an evolutionary relationship .
Small nucleolar RNAs ( snoRNAs ) and microRNAs ( miRNAs ) are two classes of abundant non-coding regulatory RNAs that carry out fundamental cellular activities but that have only been comprehensively investigated in recent years . SnoRNAs are small RNA molecules of approximately 60–300 nucleotides in length which generally serve as guides for the catalytic modification of selected ribosomal RNA nucleotides [1] , [2] . SnoRNAs associate with specific proteins , which are conserved amongst all eukaryotes , to form small nucleolar ribonucleoparticles ( snoRNPs ) . Two main groups of snoRNAs have been described . The box C/D snoRNAs , which bind the four conserved core box C/D snoRNP proteins fibrillarin , NOP56 , NOP5/NOP58 and NHP2L1 , are involved in 2′-O-ribose methylation . The box H/ACA snoRNAs , which bind the four conserved core box H/ACA snoRNP proteins DKC1 ( dyskerin ) , GAR1 , NHP2 and NOP10 , catalyse pseudouridylation . In vertebrates , most snoRNAs have been shown to reside in introns of protein coding host genes and are processed out of the excised introns [3] . However , two box C/D snoRNAs have recently been found to be transcribed from independent RNA pol II units [4] . MiRNAs are ∼18–24 nucleotide-long RNAs that are processed out of ∼70 nucleotide-long hairpin structures ( called pre-miRNAs ) [5] . In mammals , miRNAs have been shown to be involved mainly in mRNA translation inhibition [6] although recently , they have also been reported to activate translation [7] . A large class of miRNAs are encoded in introns of protein-coding genes and are co-expressed with these host genes [8]–[10] . The remaining miRNAs are encoded in independent transcription units . Some of these miRNAs have been shown to be under the control of the RNA polymerase II [11] while others are transcribed by the RNA polymerase III [12] . Many members of the snoRNA and miRNA classes are well conserved throughout evolution [1] , [2] , [13] . Correspondence between several yeast and human snoRNAs and their target sites have been established and many snoRNAs have a very high sequence identity within mammals as shown in the snoRNAbase database [14] . In the case of miRNAs , several families have been found to be well conserved in metazoans [13] , [15] . However , recent reports also suggest the existence of species- and lineage-specific snoRNAs and miRNAs [13] , [16] , [17] . These and other reports on their origin and evolution are providing clues about the emergence of large groups of these recently evolved molecules . Through bioinformatic searches , Weber [17] and Luo and Li [16] identified hundreds of human snoRNAs and snoRNA-related molecules that are derived from transposable elements ( TEs ) , thus confirming the widespread nature of this phenomenon , initially described for a small number of snoRNAs [2] , [18] . These analyses suggest that many snoRNAs result from the retroposition of existing snoRNAs that used long interspersed nuclear element ( LINE ) machinery to transpose themselves to new genomic locations . Many of these snoRNA-related molecules are surrounded by the presence of sequence features typical of retrogenes such as target site duplications ( TSDs ) and poly ( A ) tails at their 3′ end . These snoRNA retroposition events generated hundreds of sno-related molecules , termed snoRTs ( snoRNA retroposons ) by Weber [17] , many of which had never been previously identified , but some of which were previously described as functional snoRNAs [16] . SnoRNA retroposition thus not only permits maintenance of a pool of intact snoRNA copies to safeguard against the effects of deleterious mutations but could possibly also allow for the creation of regulatory RNA molecules that might bind new targets [17] . Given the stringent thresholds used to search for snoRNA copies in both studies , it is likely that many more such molecules exist in the human genome but might have diverged further from their parental copies and are yet to be discovered . Recent reports have also described some miRNAs as being derived from TEs , suggesting a possible mechanism for the rapid generation of miRNAs and their corresponding target sites . In the first such report , Smalheiser and Torvik identified six miRNAs that are derived from TEs [19] . Two subsequent studies identified a further 95 [12] and 55 [20] , [21] known miRNAs that might be derived from TEs as well as an additional 85 predicted novel TE-derived miRNA genes [20] . The TEs that are most frequently found in association with miRNAs are the L2 and MIR families [20] . As TEs are the most non-conserved sequence elements in eukaryotic genomes [22] , the generation of miRNAs through TEs represents a mechanism that could be a driving force in speciation events and evolution by rapidly creating new regulatory elements in the control of protein production [19] , [20] . A recent report investigating the small RNAs present in human cells has demonstrated the existence of specific small RNA fragments derived from larger known non-coding RNA molecules [23] . In particular , distinct small fragments of sizes between 23 and 25 nucleotides were found to map to four box H/ACA snoRNAs [23] ( listed in Table 1 ) . In addition to this , Ender and colleagues have recently reported eight box H/ACA snoRNA-derived miRNA-like molecules that can be immunoprecipitated with Ago proteins [24] . While these short H/ACA snoRNA-derived fragments might be discounted merely as non-functional degradation products , several unrelated observations suggest otherwise . Firstly , only specific fragments derived from one region of each snoRNA were identified , rather than a ladder of fragments consistent with degradation . Secondly , other snoRNAs encode smaller fragments that are stably produced . Indeed , three miRNAs present in the miRNA repository miRBase [25] can be shown to be encoded in known H/ACA snoRNAs ( listed in Table 1 ) . Although at least one pair of these miRNAs and snoRNAs are known to be co-localised in the genome as mentioned in miRBase [25] , it is not known whether the processing of these molecules is independent or dependent and sequential . Thirdly , as mentioned above , miRNA and snoRNA members have both been found to be TE-derived , suggesting a similar origin and evolution for at least some members of these small non-coding RNA classes . Here , in light of the accumulation of data suggesting a connection between box H/ACA snoRNAs and miRNA-like molecules , we investigate the possibility of an evolutionary relationship between members of these classes of RNA .
A comparison between the genomic positions of reported miRNA genes from miRbase [25] and box H/ACA snoRNAs reveals three occurrences of overlap between these RNA species ( Table 1 , top section ) . In all three cases , between 75% ( mir-1291 ) and 97% ( mir-1248 ) of the miRNA hairpin is contained within the snoRNA ( using the coordinates of the UCSC Genome Browser as described in the Methods ) . Moreover , in all three cases , greater than 90% of the mature miRNA as defined in miRbase release 11 . 0 [25] is contained within the snoRNA . In addition to these known miRNAs encoded in box H/ACA snoRNAs , ten small fragments matching exactly to portions of eleven box H/ACA snoRNAs have been detected [23] , [24] and are listed in Table 1 ( bottom section ) . One of the fragments is identical to two very similar H/ACA snoRNAs , ACA7 and ACA7B . Of these ten small fragments , seven have been shown to be bound by Ago proteins and one of these , ACA45 sRNA , has experimentally validated targets [24] . Apart from HBI-100 , all box H/ACA snoRNAs from Table 1 that contain experimentally detected smaller fragments are either experimentally verified snoRNAs or close paralogues of such experimentally validated snoRNAs . ACA34 , ACA45 , ACA47 , ACA56 , ACA3 , ACA50 , ACA7 , HBI-61 , U17b , U71a and U92 have been shown experimentally to display characteristics of H/ACA snoRNAs [26]–[31] . ACA36B is a close paralogue ( 88% identity ) of the experimentally validated H/ACA snoRNA ACA36 [28] . ACA7B is a close paralogue of ACA7 ( 98% identical ) . ACA7B and ACA36B share both their predicted rRNA targets with ACA7 and ACA36 respectively , as described in the snoRNAbase [14] . In addition , as shown in Table 1 , seven of the box H/ACA snoRNAs that encode smaller experimentally detected fragments share their predicted rRNA and snRNA targets with other box H/ACA snoRNAs and one ( U17b ) does not have a known target . The remaining six box H/ACA snoRNAs , HBI-61 , ACA45 , ACA47 , ACA56 , ACA3 and U92 , are predicted to guide the pseudouridylation of known modified residues [14] and no other snoRNA is known to serve as a guide for these residues . The UCSC Genome Browser mammalian conservation track shows that for all snoRNAs listed in Table 1 except U71a and ACA56 , the conserved region around these molecules covers the entire snoRNA molecules , not only the miRNA hairpins or small RNA fragments detected ( Figure 1 and Figure S1 ) . The miRNA hairpins of miR-664 and miR-1291 have short 3′ and 5′ regions respectively that do not overlap with a snoRNA . These regions correspond to the least well conserved regions of the whole miRNA/snoRNA molecules . This suggests that these regions originally encoded snoRNAs and not necessarily miRNAs in the most recent common ancestor . Indeed , to our knowledge , apart from mir-664 and ACA45 sRNA , none of the other miRNAs and smaller fragments have been detected in other mammalian species , suggesting the capability of generating smaller RNA molecules from these snoRNAs might be a recent event . The box H/ACA snoRNA/miRNA relationship described above was further investigated by studying all known miRNAs to determine whether they might be encoded within genomic regions predicted to harbour H/ACA snoRNAs . Indeed , if some miRNAs have evolved from H/ACA snoRNA encoding regions , they might still display snoRNA features . The mammalian version of the snoGPS program predicts pseudouridylation guides in human , mouse and rat genomes by scoring weakly conserved primary and secondary structure motifs using a deterministic search algorithm [32] . The mammalian version of the snoGPS program also includes a cross-species implementation ( snoGPS-C ) which takes account of conservation between several mammalian genomes to predict box H/ACA snoRNAs [32] . A locally-installed copy of the mammalian snoGPS program was used to scan with the two-hairpin model for the presence of box H/ACA snoRNAs in 676 distinct sequences consisting of human miRNAs from miRBase ( version 11 . 0 ) [25] and an additional 175 padding nucleotides upstream and downstream ( referred to as the extended miRNA molecules ) . We did not use the cross-species implementation of snoGPS ( snoGPS-C ) because many of the newly described snoRNAs ( especially the TE-derived snoRNAs ) are lineage- or species-specific [17] . In order to investigate whether the number of snoGPS predicted hits above a certain threshold was significant , we used snoGPS to scan 100 sets of 676 randomly generated sequences of same length distribution as the miRNAs under study , as described in the Methods section . The number of hits above a given threshold for both the set of extended miRNA sequences under study and the randomly generated sequences is shown in Figure 2 . MiRNA precursors , like H/ACA snoRNAs , consist of at least one hairpin . To control for this , a second set of control sequences consisting of 676 randomly generated hairpins of same length distribution and minimum free-energy distribution ( as calculated by RNAfold [33] ) as the miRNA hairpins under study was generated . 100 such random hairpin sets were scanned using snoGPS and their average number of hits is shown in Figure 2 . As expected , the number of hits for the random hairpin groups is significantly higher than the number of hits for the random sequence groups that have not been constrained to form hairpins . However , for all hit score thresholds investigated , the number of hits predicted by snoGPS for the miRNA group is significantly higher than the number of hits predicted for any of the randomly generated groups . This suggests that genomic regions around a significant number of miRNAs contain features that very closely resemble box H/ACA snoRNAs . 148 distinct extended miRNA molecules were predicted to encode at least one hit above a score of 35 . 0 , which is the threshold that is ‘typically’ used when predicting new candidate snoRNAs by snoGPS-C [32] . Since we chose to use the original snoGPS version , we set the threshold higher , at 40 . 0 , in order to consider only very likely candidates . Taking this conservative threshold , 29 distinct extended miRNA molecules were predicted to encode at least one hit . All predicted hits were folded to reveal their predicted secondary structure , using RNAstructure [34] . When the highest snoGPS hit could not be folded in a secondary structure that was within 10% of the lowest RNAstructure predicted minimum free-energy , snoGPS hits of lower score ( but still above 40 ) were considered . The best snoGPS hit is defined as the snoGPS hit with highest score that has a predicted secondary structure minimum free-energy within 10% of the lowest predicted minimum free-energy structure for this molecule . Twenty extended miRNA regions had best snoGPS hits above a score of 40 . 0 and the remaining nine extended miRNA regions with lower best snoGPS hits were not considered further . Table 2 describes the best hit for each of these twenty extended miRNA molecules . The position of the predicted H/ACA snoRNA was compared to the position of the miRNA hairpin , taking coordinates downloaded from the UCSC Table Browser as described in the Methods . Apart from the predicted snoRNAs that contain mir-151 and mir-215 , all predicted snoRNAs in Table 2 contain at least 90% of their encoded miRNA hairpins . Approximately 80% of the hairpins of both mir-151 and mir-215 are contained in their respective predicted snoRNA . In addition , for all miRNAs described in Table 2 , at least 90% of the mature miRNA is contained within the predicted snoRNA . In this respect , all these snoRNA-predicted miRNA pairs are similar to the known H/ACA snoRNAs that encode smaller fragments detected experimentally , described in Table 1 . SnoGPS predicts guide sequences and corresponding rRNA pseudouridylation sites within snoRNAs . For all the best snoGPS hits with scores above 40 listed in Table 2 , the predicted pseudouridylation sites are reported . While most of these predicted pseudouridylation sites are known to be recognised by already reported box H/ACA snoRNAs , four are labelled as having an unknown guide in snoRNAbase [14] . Indeed , the H/ACA snoRNAs predicted in the extended region around mir-549 , mir-140 , mir-1262 and mir-605 are all predicted to serve as guides for experimentally validated pseudouridylation sites whose guides are unknown , making these interesting candidates for further studies . Some of these might represent genomic regions with a dual function , serving both to produce miRNAs and snoRNAs . The miRNAs reported in miRbase have not all been validated to the same extent . While the mature forms of some of the miRNAs have only been identified with a very small number of sequence reads , others have been identified by larger numbers of reads , display characteristic miRNA signatures ( with detection of a much smaller number of star reads than the mature form reads [24] , [35] ) and have been functionally validated . For each of the twenty miRNAs described in Table 2 , we include the number of sequence reads and when available , the number of star reads , as reported in the literature . Ten of the miRNAs in Table 2 have been identified with at least 10 reads and four of these ( miR-151 , miR-885 , miR-140 and miR-520a ) also have corresponding star reads of lower abundance . On the other hand , three reported miRNAs with snoRNA-like features , miR-549 , miR-548m and miR-605 , have been identified with fewer than 4 reads . While the best snoGPS hit has been investigated here , it is important to point out that some extended miRNA regions obtain more than one high-scoring hit . Most notably , mir-548d-1 and mir-548d-2 have high-scoring hits in both their hairpins , in a manner reminiscent of well validated H/ACA snoRNAs such as E2 and U65 . Box H/ACA snoRNAs have very distinct features . They usually consist of two hairpins , each of which is followed by short single-stranded regions ( the H and ACA boxes ) . While the H box is located between the two hairpins , the ACA box is located at the 3′ end of the molecule . One or both of the hairpins contain bulges , allowing base-pairing with the target RNA , in complex pseudo-knot structures . In order to better characterise the predicted snoRNAs encoding miRNAs and visualise the position of the mature miRNA within these molecules , all predicted snoRNA sequences were folded using RNAstructure [34] and are shown in Figure 3B and Figure S2 . In addition , the predicted secondary structure of the four snoRNAs encoding known miRNAs ( from Table 1 ) are also shown ( Figure 3A ) . Most of the predicted snoRNAs encoding miRNAs resemble typical snoRNAs with two main hairpins , characteristic boxes and one or two bulges containing the predicted RNA target complementary sites Because numerous snoRNAs and miRNAs have been described as being derived from TEs , all extended miRNA molecules predicted to have box H/ACA snoRNA features surrounding them ( from Table 2 ) were further investigated for the presence of repeat elements using RepeatMasker ( http://www . repeatmasker . org ) . Sixteen of the twenty miRNAs originally considered have repeat elements either overlapping the predicted snoRNA encoding the miRNA or within 400 nucleotides . The position of the repeat elements with respect to the position of the miRNA and predicted snoRNA is shown in Figure 4 and Figure S3 . In addition , putative L1 consensus recognition sites and flanking target site duplications ( TSDs ) , which are characteristic of retrogenes , were also identified surrounding many of these molecules ( Figure 4 and Figure S4 ) . Some of these putative snoRNA-encoded miRNA regions have a genomic structure that is very similar to numerous snoRTs [16] , consisting of the snoRNA/miRNA region in close proximity to a downstream SINE member repeat element and flanked by target site duplications ( TSDs ) . In addition , immediately upstream from the 5′ TSD , an L1 consensus recognition site is often found and a poly ( A ) tail can be identified upstream from the 3′ TSD . Three such examples resembling the HBI-61c snoRT from [16] are shown in Figure 4A–C . Shown in Figure 4D is the genomic region surrounding mir-605 , which consists of two pairs of TSDs , one of which flanks a SINE repeat element and the other of which flanks the whole snoRNA/miRNA/SINE region . This structure resembles the ACA18e snoRT example from [16] . The examples shown in Figure 4 suggest that it is the predicted snoRNA and not the miRNA hairpin that was captured in the retrogene construct and initially transposed , in a manner similar to snoRT events previously described [16] , [17] . To explore further potential snoRNA-like features of these miRNA precursors and to investigate whether they have retained some H/ACA snoRNA functionality , we tested whether they can bind dyskerin , a protein component of the functional box H/ACA snoRNPs . Dyskerin serves as the pseudouridine synthase [36] and is proposed to bind the ACA box [37] . Five of the twenty snoRNA-like miRNAs , mir-664 , mir-151 , mir-605 , mir-215 and mir-140 were selected for this analysis because they are expressed in HeLa cells . Purified nuclei from HeLa cells expressing YFP-dyskerin or GFP as a control were immunoprecipitated using an antibody against the fluorescent proteins as described in the Methods . The RNA isolated from these samples was analysed by RT-PCR for the presence of the molecules of interest . As shown in Figure 5C , in addition to E2 ( a well-characterised box H/ACA snoRNA ) , five snoRNA-like miRNA precursors ( the extended regions of mir-664 , mir-151 , mir-605 , mir-215 and mir-140 ) are bound by dyskerin . In contrast , the precursor of hsa-let-7g , a miRNA precursor with no similarity to box H/ACA snoRNAs is not pulled down by dyskerin . And as expected , other abundant nuclear RNAs including GAPDH pre-mRNA , the small nuclear RNA U1 , the box C/D snoRNA U3 and 5S rRNA are not immunoprecipitated by dyskerin . These binding experiments confirm the in silico predictions that some miRNA precursors sufficiently resemble box H/ACA snoRNAs to be bound by dyskerin . Two of the dyskerin-bound miRNA precursors were further characterised by fractionated northern analysis to investigate where the predicted snoRNAs and smaller fragments localise in the cell . As shown in Figure 6 , bands of the size of the predicted H/ACA snoRNA full-length molecules encoding mir-151 and mir-664 localise to the nucleolus ( bands labelled with ‘a’ in panels 6A and 6B ) . Bands of the size of the miRNA hairpins are detected in all three fractions although the putative mir-151 hairpin is mainly nucleolar whereas the putative mir-664 hairpin accumulates predominantly in the nucleoplasm and cytoplasm ( bands labelled with ‘b’ in panels 6A and 6B ) . The mir-151 mature form is also detected and mainly found in the nucleoplasm and cytoplasm ( bands labelled with ‘c’ in panel 6A . For a longer exposure , please see Figure S5 ) . And a band slightly larger than the mir-664 mature form localises mainly in the nucleoplasm .
Numerous miRNAs have been previously shown to be repeat-derived [12] , [19]–[21] and many snoRNAs have been described as retrogenes [16] , [17] . Here , we hypothesize that some reported miRNAs have evolved from box H/ACA snoRNAs or snoRTs . Several lines of evidence support this possibility . Fourteen known box H/ACA snoRNAs encode smaller fragments of miRNA size that have been experimentally detected , three of which are reported miRNAs . Analysis of mammalian conservation patterns suggests that these genomic regions originally encoded the full-length H/ACA snoRNA molecules and not only the miRNAs . If a subgroup of miRNAs has indeed evolved from box H/ACA snoRNAs , we reasoned that although some of these miRNAs might have sufficiently evolved to no longer bear measurable similarity to H/ACA snoRNAs , others might display detectable H/ACA snoRNA features . In an effort to further characterise the prevalence of the relationship between these two classes of small RNA molecules , we scanned the regions encoding known miRNAs for the presence of box H/ACA snoRNA features using the snoGPS predictor . We identified twenty reported miRNAs from miRBase [25] that are encoded in larger regions predicted with high scores to be box H/ACA snoRNAs . The predicted box H/ACA snoRNAs display usual box H/ACA snoRNA features and resemble the fourteen box H/ACA snoRNAs that encode experimentally detected smaller fragments . In addition , the genomic sequence surrounding several of the predicted snoRNA-like miRNAs very closely resembles those described for some snoRTs [16] . These analyses show that some genomic regions previously reported to encode miRNAs resemble regions that encode H/ACA snoRNAs on numerous levels . This suggests that these miRNAs have evolved from H/ACA snoRNAs or snoRTs . We applied stringent selection criteria in our analysis , so anticipate that other box H/ACA snoRNA-like miRNA precursors also exist but have not been identified here . Due to the inherent similarity between miRNAs and snoRNAs , such a relationship is easy to overlook as once a region is categorized as belonging to one molecular class , it is often no longer considered when searching for other types of molecules . The human genome has been scanned previously for the presence of box H/ACA snoRNAs using mammalian snoGPS [32] . However , the search space was limited to the 20% most well conserved regions between the human , mouse and rat genomes . In addition , the dataset was repeat-masked , thus eliminating repeat-derived regions such as those encoding many miRNAs . Finally , the dataset was restricted to sequences that do not overlap with known features in the UCSC Human Genome Browser database , thus probably eliminating all known miRNAs . As a consequence , it is not surprising that no miRNA encoding regions were identified as also encoding predicted snoRNAs . Moreover , at least one recent snoRNA predictor , SnoReport [38] uses miRNAs as negative training examples , thus making it very unlikely to identify any of the snoRNA-like miRNA regions described here . Although no significant sequence similarity is detected between predicted snoRNA molecules encoding miRNAs and the known snoRNAs that target the same pseudouridylation sites , it is interesting to note that three snoRNAs ( ACA52 , HBI-61 and ACA19 ) share their target pseudouridylation sites with eight of the predicted snoRNAs encoding miRNAs ( Table 2 ) . This situation also exists amongst known snoRNAs , some of which share the same target site without displaying significant sequence similarity . In particular , examples exist of a snoRNA harbouring two guide regions , each of which is shared with a different snoRNA . For example , ACA22 which shares one of its targets with ACA33 and the other with U64 , has no significant sequence similarity with either molecule . Similarly , ACA50 shares target sites with ACA36 , ACA8 and ACA62 but while it has high sequence identity with ACA62 , it has no significant sequence similarity with ACA8 or ACA36 . This redundancy in rRNA complementarity may suggest some box H/ACA snoRNAs and snoRTs are not under as much selective pressure to avoid mutations . We hypothesize that some of these snoRNA encoding regions might be in the process of evolving from functional snoRNAs to miRNA-like precursors . This process might be facilitated by the fact that box H/ACA snoRNAs have a structure ( 2 hairpins ) that is probably favourable to the formation of miRNAs . The in silico data presented here support these ideas . We tested the predictions by experimentally showing that the precursors of five of the predicted snoRNA-like miRNAs , mir-664 , mir-151 , mir-605 , mir-215 and mir-140 , interact with dyskerin , a protein component of functional H/ACA snoRNPs . While a lack of interaction to dyskerin could not rule out an evolutionary relationship between these miRNAs and snoRNAs as the miRNAs might have evolved sufficiently to no longer interact with functional protein components of snoRNPs , the detection of such an interaction considerably reinforces such claims . These results show that the snoRNA-like miRNA precursors sufficiently resemble box H/ACA snoRNAs to bind dyskerin , strengthening the possibility of an evolutionary relationship between these molecules . Further experiments will be necessary to investigate whether these molecules also retain the capability of targeting rRNA in vivo . It will also be necessary to experimentally test whether the remaining fifteen predicted snoRNA-like miRNA precursors also display aspects of H/ACA snoRNA functionality . The fact that three of these molecules ( mir-548d-1 , mir-1297 and mir-616 ) display the sequence AGA instead of the canonical ACA box could indicate that they have evolved sufficiently to no longer retain H/ACA snoRNA functionality . We do note that while identifying molecules that display both miRNA and snoRNA functionality supports our evolutionary hypothesis , we also expect to find a larger number of molecules that display features of both molecules but do not represent completely prototypical examples . It is interesting to note that Saccharomyces cerevisiae has snoRNAs but no reported miRNAs , consistent with a relationship where primordial snoRNAs may have given rise to certain classes of miRNAs . This idea is supported by a recent article by Saraiya and Wang reporting that the primitive parasitic protozoan Giardia lamblia , which does not have RNA interference capabilities but has miRNA processing machinery , uses box C/D snoRNAs as miRNA precursors [39] . In addition to this , Taft and colleagues have recently reported that most snoRNAs in animals , Arabidopsis and Schizosaccharomyces pombe generate small RNAs ( of ∼20–24 nucleotides in length for animal box H/ACA snoRNAs ) , which are associated with argonaute proteins [40] . Current data such as these dual function molecules with both miRNA and snoRNA capabilities which exist in both human [24] and Giardia [39] and likely many other organisms [40] suggest this process of evolving from a snoRNA encoding genomic region to a miRNA-like encoding region could be ongoing . In addition to investigating whether some of the H/ACA snoRNA-like miRNA precursors display functional H/ACA snoRNA capability by binding to dyskerin ( Figure 5 ) , we have also characterised the cellular localisation of two of these molecules: the precursors of mir-151 and mir-664 ( Figure 6 ) . While bands of the size of the predicted full-length H/ACA snoRNA molecules localise to the nucleolus , consistent with their binding to dyskerin , bands of the size of the predicted hairpin form of these miRNAs can be found in all three fractions considered but accumulate mainly in the nucleolus ( in the case of mir-151 ) and in the nucleoplasm and cytoplasm ( in the case of mir-664 ) . The mature form of mir-151 accumulates mainly in the nucleoplasm which is unusual for a miRNA but might be a consequence of the snoRNA features displayed by its precursor . These results are consistent with a recent study showing that the precursors and/or mature form of a number of rat miRNAs accumulate in the nucleolus [41] . Further studies will be required to investigate the exact nature and role of each of these molecules in these cellular compartments as well as how they are processed . While all the miRNAs characterised in Table 2 are classified as miRNAs in miRBase [25] , they have not all been extensively analysed . Of the five extended miRNA regions that we experimentally found to be bound by dyskerin , three ( mir-151 , mir-140 and mir-215 ) have been further characterized and functionally validated , either by studies of their processing into their mature form or validation of their targets and effects . Indeed , mir-151 has been shown to be processed into its mature form by usual miRNA processing machinery [42] while functional targets of mir-140 have been experimentally validated [43] . And mir-215 , which has been shown to have reduced expression in cancer tissues compared to normal cells , is capable of inducing cell-cycle arrest , colony suppression and cell detachment from a solid support when transfected into cells [44] . Mir-664 and mir-605 have not , to our knowledge , been further functionally validated and will require additional experimental evidence to confirm they are true miRNAs . In particular , mir-605 has only been identified previously with one sequence read [45] . Given that the extended region of mir-605 is predicted to serve as a guide for an experimentally validated pseudouridylation site whose guide is unknown , we postulate that this region encodes an H/ACA snoRNA rather than a miRNA . This type of analysis can thus be used to filter out unlikely miRNA candidates from the large miRNA repositories which contain many poorly characterized molecules . A recent large-scale study defining an expression atlas for mammalian miRNAs , by Landgraf and colleagues , classifies known miRNAs into four different groups: prototypical , repeat-derived , repeat-clustered and unclassified [46] . A lack of repetitiveness , evolutionary conservation and 5′ end processing were considered to classify miRNAs as prototypical . Only two ( mir-215 and mir-140 ) of our twenty miRNAs encoded in predicted snoRNAs are classified as protypical in this study . The remaining eighteen miRNAs were either classified as repeat-derived ( 5 miRNAs ) , repeat-clustered ( 1 miRNA ) , unclassified ( 2 miRNAs ) or were not considered in the study ( 10 miRNAs ) . The results of a recent deep-sequencing study [45] analysed in the context of the Landgraf study reveals that only 17% of all miRNAs identified in this manner are prototypical [46] . The remaining miRNAs displayed irregularities in their processing or unusual sequence conservation patterns [46] . The authors further went on to investigate possible functional implications , determining that unlike non-prototypical miRNAs , prototypical miRNAs showed enrichment of putative target sites in 3′UTRs [46] . In light of the relationship uncovered here between box H/ACA snoRNAs and miRNAs , it is reasonable to speculate that snoRNA-encoded miRNAs might be in the process of evolving into functional miRNAs but still retain some non-miRNA-like features which could preclude them from being classified as prototypical . Alternatively , perhaps these molecules should be classified as a separate small RNA class . Further analysis will be necessary to clarify the role and exact relationship of these molecules .
The genomic positions , sequences and flanking regions of human miRNA genes and box H/ACA snoRNA genes were downloaded from the UCSC Table Browser [47] , wgRNA table [14] , [25] using the March 2006 assembly of the human genome . Repeat elements surrounding these genomic regions were identified with the RepeatMasker program ( http://www . repeatmasker . org ) . MiRNA and box H/ACA snoRNA genes as well as repeat-elements and sequence features were visualised with the UCSC Genome Browser [48] , using information from the wgRNA table [14] , [25] , the Vertebrate Multiz Alignment and PhastCons Conservation utilities [49] , [50] as well as custom tracks . 676 distinct human miRNA encoding regions ( including the miRNA hairpins and 175 flanking nucleotides on either side ) were scanned for the presence of box H/ACA snoRNAs using a locally-installed copy of the snoGPS program [32] . In addition to known miRNAs , two sets of control sequences were also scanned using snoGPS . The first sets of control sequences consisted of 676 randomly generated sequences having the same nucleotide composition as human intronic sequences and the same length distribution as the miRNA encoding regions . The second sets of control sequences consisted of randomly generated hairpins having the same length distribution and minimum free-energy distribution as the miRNA hairpin set . RNAfold [33] was used to predict the minimum free-energy of the randomly-generated hairpin sequences . The snoGPS parameters used throughout this analysis were as follow . The target sites for pseudouridylation were defined in the file ‘snAndRrna . targ’ provided with the snoGPS download . The descriptor and scoretable files used were respectively ‘MamGUs2 . v3 . desc’ and ‘human . v3 . scoretables’ , provided with the snoGPS download . RNA secondary structures were predicted using RNAstructure 4 . 5 [34] and annotated using RnaViz 2 . 0 [51] . Immunoprecipitations were prepared as previously described [52] . Nuclear lysates were prepared from HeLaYFP-Dyskerin and HeLaGFP stable cell lines . Purified nuclei were resuspended in RIPA buffer to solubilise proteins . Fluorescence proteins were immunoprecipitated using GFP binder ( ChromoTek ) covalently coupled to NHS-activated Sepharose 4 Fast Flow beads ( GE Healthcare at 1 mg/ml as previously described [53] ) . Samples were divided in two and for Input samples RNA was isolated from one half of each nuclear lysate . RNA was isolated by the TRIzol method with DNase I treatment , according to manufacturer's instructions ( Invitrogen ) . RT-PCR was performed to detect immunoprecipitated RNAs . Reverse transcription and PCR were performed with the following gene-specific primers ( hsa-pri-let-7g: 5′-CGCTCCGTTTCCTTTTGCCTG-3′ and 5′ TACAGTTATCTCCTGTACCGG-3′ , U3: 5′-AGAGGTAGCGTTTTCTCCTGAGCG-3′ and 5′ ACCACTCAGACCGCGTTCTC-3′ , pre-GAPDH: 5′-CGCATCTTCTTTTGCGTCGCCAG-3′ and 5′-GGTCAATGAAGGGGTCATTGATGGC-3′ , U1: 5′-TACCTGGCAGGGGAGATACCATGATC-3′ and 5′-GCAGTCGAGTTTCCCACATTTGGGG-3′ , 5S: 5′-ACGCGCCCGATCTCGTCTGAT-3′ and 5′-GCCTACAGCACCCGGTATTCCC-3′ , miR-664: 5′-GTGTTAAGTTCAGTTCAGGGTAG-3′ and 5′-CATTTTGTAGGCTGGGGATAAATG-3′ , miR-151: 5′-GGCTTACCCTATGCTGCTATA-3′ and 5′-GTAGGGGATGAGACATACTAGAC-3′ , miR-605: 5′-CTGGTCTTGAACTCCTGATCTC-3′ and 5′- GCTGTCAGCCTGTAACATAGG-3′ , miR-215: 5′-CCAAAAAGATCCAATAATGGAAGAGGATTAAAG-3′ and 5′-TTGAAGTAGCACAGTCATACAG-3′ , miR-140: 5′-GTGTGTCTCTCTCTGTGTCC-3′ and 5′-GGATGTCCCAAGGGGGCCAG-3′ ) using the SuperScript one-step RT-PCR kit ( Invitrogen ) . To decide linearity of cycles , we performed real time PCR using the Superscript III Platinum SYBR Green one-step qRT-PCR Kit ( Invitrogen ) and Rotor-Gene RG-3000 system ( Corbett Research ) . The same amount of RNA for input and immunoprecipitated RNA ( IP ) was used as templates for RT-PCR reactions . Each experiment was repeated three times independently . HeLa cell extracts were fractionated using sucrose gradients , as previously described [54]–[56] . Total HeLa cell RNA and RNA from separate cytoplasmic , nucleoplasmic and nucleolar fractions was isolated using the TRIzol method , with Dnase I treatment , according to manufacturer's instructions ( Invitrogen ) . Equal amounts of RNA from each sample were separated by 8M Urea polyacrylamide denaturing gel electrophoresis in 1×MOPS buffer and the RNA transferred onto nylon membrane ( Hybond-N; Amersham ) by electro blotting . After chemical cross linking , the membrane was hybridized with 32P 5′ end-labelled oligoribonucleotide probes specific for the following RNA species; ( mir-664: 5′- UGUAGGCUGGGGAUAAAUGAAUA-3′ , mir-151: 5′-CCUCAAGGAGCUUCAGUCUAG-3′ , tRNA-Ile 5′-UGGUGGCCCGUACGGGGAUCGA-3′ , U11: 5′-TCTTGATGTCGATTCCGCACGCAGAGCAATCGAGTTGCCC-3′ and U3: 5′-CACTCAGACCGCGTTCTCTCCCTCTCACTCCCCAATACGG-3′ ) . High sensitivity RNA blots were prepared as previously described [57] . | The major functions known for RNA were long believed to be either messenger RNAs , which function as intermediates between genes and proteins , or ribosomal RNAs and transfer RNAs which carry out the translation process . In recent years , however , newly discovered classes of small RNAs have been shown to play important cellular roles . These include microRNAs ( miRNAs ) , which can regulate the production of specific proteins , and small nucleolar RNAs ( snoRNAs ) , which recognise and chemically modify specific sequences in ribosomal RNA . Although miRNAs and snoRNAs are currently believed to be generated by different cellular pathways and to function in different cellular compartments , members of these two types of small RNAs display numerous genomic similarities , and a small number of snoRNAs have been shown to encode miRNAs in several organisms . Here we systematically investigate a possible evolutionary relationship between snoRNAs and miRNAs . Using computational analysis , we identify twenty genomic regions encoding miRNAs with highly significant similarity to snoRNAs , both on the level of their surrounding genomic context as well as their predicted folded structure . A subset of these miRNAs display functional snoRNA characteristics , strengthening the possibility that these miRNA molecules might have evolved from snoRNAs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"molecular",
"biology/rna-protein",
"interactions",
"evolutionary",
"biology/genomics",
"evolutionary",
"biology/bioinformatics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics"
] | 2009 | Human miRNA Precursors with Box H/ACA snoRNA Features |
Pseudomonas aeruginosa is a Gram-negative pathogen that can lead to severe infection associated with lung injury and high mortality . The interleukin ( IL ) -36 cytokines ( IL-36α , IL-36β and IL-36γ ) are newly described IL-1 like family cytokines that promote inflammatory response via binding to the IL-36 receptor ( IL-36R ) . Here we investigated the functional role of IL-36 cytokines in the modulating of innate immune response against P . aeruginosa pulmonary infection . The intratracheal administration of flagellated cytotoxic P . aeruginosa ( ATCC 19660 ) upregulated IL-36α and IL-36γ , but not IL-36β , in the lungs . IL-36α and IL-36γ were expressed in pulmonary macrophages ( PMs ) and alveolar epithelial cells in response to P . aeruginosa in vitro . Mortality after bacterial challenge in IL-36 receptor deficient ( IL-36R-/- ) mice and IL-36γ deficient ( IL-36γ-/- ) mice , but not IL-36α deficient mice , was significantly lower than that of wild type mice . Decreased mortality in IL-36R-/- mice and IL-36γ-/- mice was associated with reduction in bacterial burden in the alveolar space , bacterial dissemination , production of inflammatory cytokines and lung injury , without changes in lung leukocyte influx . Interestingly , IL-36γ enhanced the production of prostaglandin E2 ( PGE2 ) during P . aeruginosa infection in vivo and in vitro . Treatment of PMs with recombinant IL-36γ resulted in impaired bacterial killing via PGE2 and its receptor; EP2 . P . aeruginosa infected EP2 deficient mice or WT mice treated with a COX-2-specific inhibitor showed decreased bacterial burden and dissemination , but no change in lung injury . Finally , we observed an increase in IL-36γ , but not IL-36α , in the airspace and plasma of patients with P . aeruginosa-induced acute respiratory distress syndrome . Thus , IL-36γ and its receptor signal not only impaired bacterial clearance in a possible PGE2 dependent fashion but also mediated lung injury during P . aeruginosa infection .
Pseudomonas aeruginosa is a Gram-negative bacterium that causes acute nosocomial infection as well as chronic infection in immunocompromised hosts . Infection with P . aeruginosa can lead to sepsis , pneumonia , and lung injury , which are often severe and life threatening [1] . Poor clinical outcomes in this disease are believed to be due to virulence factors expressed by this pathogen , increasing rate of multidrug resistance in P . aeruginosa , and the immune status of the infected hosts [2 , 3] . Interleukin ( IL ) -36 cytokines , including three agonists IL-36α , IL-36β , IL-36γ and an antagonist IL-36Ra , are recently described members of IL-1 family of cytokines . IL-36 agonists bind the same receptor complex , consisting of the IL-36 receptor ( IL-36R ) and IL-1 receptor accessory protein ( IL-1RAcP ) , which is shared with the IL-1 receptor and the IL-33 receptor [4] . IL-36R agonists are expressed by stimulated immune cells , such as monocytes and macrophages , dendritic cells , and epithelial cells [5–7] . The IL-36 receptor ligands induce pro-inflammatory cytokine and chemokine expression and contribute to neutrophil accumulation , dendritic cell activation and polarization of T helper 1 and IL-17 producing T cells [4 , 8 , 9] . Intratracheal administration of IL-36α or IL-36γ in mice induces a rapid influx of neutrophils into the lungs and pro-inflammatory cytokines and chemokines [10 , 11] . Innate immune cell recruitment and phagocytic bacterial clearance , including neutrophils and macrophages , have critical roles in the host defense during the early stage of P . aeruginosa infection [12 , 13] . Moreover , IL-36γ mRNA is upregulated in human bronchial cells after infection with P . aeruginosa [14] . Taken together , these observations suggest that IL-36 cytokines may play an important role in host defense against P . aeruginosa , perhaps by contributing to inflammatory cell recruitment/activation during infection , To date , the role of IL-36 cytokines in the host defense of P . aeruginosa infection has not been defined . Prostaglandin E2 ( PGE2 ) is a major product of arachidonic acid metabolism and its production is dependent on the cyclooxygenases ( COX-1 and COX-2 ) . Whereas COX-1 is expressed constitutively in most of cells and believed to be required for immune homeostasis , COX-2 is primarily an inducible enzyme that is expressed in response to stimulation by pro-inflammatory cytokines ( e . g . TNF-α and IL-1β ) , microbial pathogens , and endogenously produced growth factors [15] . Cytotoxic strains of P . aeruginosa induce the production of PGE2 in pulmonary macrophages through COX-2 activation [16] . In vitro studies demonstrated that PGE2 can impair phagocytic properties and bactericidal activity of alveolar macrophages during P . aeruginosa infection [17 , 18] . Conversely , COX-2 inhibition or genetic deletion have been shown to enhance bacterial clearance and reduce mortality in a murine P . aeruginosa murine model [16 , 19] . No studies have assessed possible cross-talk between IL-36 cytokines and eicosanoids such as PGE2 . We hypothesized that IL-36 cytokines and their receptor; IL-36R regulate host mucosal immunity in acute P . aeruginosa lung infection . In this study , we demonstrate that IL-36γ produced by pulmonary macrophages ( PMs ) and alveolar epithelial cells ( AECs ) during P . aeruginosa lung infection promoted deleterious effects on host outcome . These deleterious effects appear to be mediated , in part , by IL-36γ-induced production of PGE2 , resulting in impaired bacterial clearance , and IL-36γ-driven lung injury that occurred in a fashion independent of PGE2 .
To examine whether IL-36 cytokines are expressed in P . aeruginosa lung infection , we first measured IL-36 cytokine mRNA in the lungs of wild-type ( WT ) C57B/6 mice infected with a flagellated cytoxic strain of P . aeruginosa ( ATCC 19660 ) . Both IL-36α and IL-36γ mRNA levels were significantly elevated in P . aeruginosa infected lungs at 6 h and 24 hrs post bacterial challenge ( Fig 1A ) . No IL-36β mRNA was detected in P . aeruginosa infected lungs . Whereas the production of IL-36α in BAL fluid peaked at 4h and returned to baseline by 24 h after P . aeruginosa administration , IL-36γ levels in BAL fluid progressively increased at the 24 hr time point ( Fig 1B ) . Similarly , both IL-36α and IL-36γ protein levels in homogenized lung tissue were increased in P . aeruginosa infection ( Fig 1C ) . Considerably higher quantities of IL-36γ were found in BAL fluid and whole lung homogenates as compared to IL-36α ( 5–20 fold ) Also , the expression of these cytokines was compartmentalized , as no IL-36α and IL-36γ was detected in plasma at these time points . Respiratory epithelial cells and pulmonary macrophages are primary innate immune cells involved in pulmonary bacterial infection . To determine cellular source ( s ) of IL-36α and IL-36γ during P . aeruginosa infection , primary PMs and AECs isolated from WT mice were treated with lipopolysaccharide ( LPS; 1μg/ml ) or a multiplicity of infection ( MOI ) 10 of live or heat-killed P . aeruginosa . Significant induction of IL-36α and IL-36γ mRNA was observed in PMs in response to LPS , live or heat-killed P . aeruginosa as early as 4 h post stimulation , with persistent expression of IL-36γ mRNA level to 24h ( Fig 2A ) . The expression levels of IL-36α and IL-36γ mRNA were significantly elevated in AECs treated with live P . aeruginosa at both 4 h and 24 h after stimulation ( Fig 2B ) . While induction of IL-36α and IL-36γ mRNA levels was similar in PMs treated with either live or HK P . aeruginosa , live bacteria lead to much greater induction of IL-36α and IL-36γ mRNA in AECs , as compared to heat-killed bacteria ( Fig 2A and 2B ) . We next examined secretion of IL-36α and IL-36γ protein from P . aeruginosa treated-PMs and AECs . Previous studies indicated that extracellular adenosine triphosphate ( ATP ) [20 , 21] , or Caspase-3/7 activation [7 , 22] , were required for extracellular secretion of IL-36 cytokines , suggesting non-classical secretion mechanisms . Specifically , activation of the P2X7 receptor by ATP stimulation leads to changes in the morphology of lung epithelial cells [23] and macrophages [24] , including plasma membrane blebbing , microvesicle release , and ultimately to apoptosis . Johnston et . al . demonstrated that secretion of IL-36γ in keratinocytes in response to bacterial flagellin was dependent on co-stimulation with ATP [21] . Marin et . al . observed that co-stimulation with LPS and ATP is necessary for IL-36α secretion from bone marrow-derived macrophages [20] . After 24 h in culture , ATP ( 5mM ) was added , and conditioned medium was collected 20 min after ATP stimulation . In the absence of ATP , the secretion of IL-36α and IL-36γ into conditioned media ( CM ) was not elevated in LPS , live and heat-killed P . aeruginosa-treated PMs and AECs . Though we found a significant increase of IL-36α and IL-36γ production by PMs in response to live or heat-killed P . aeruginosa in combination with ATP ( Fig 2C ) , only live bacteria plus ATP treatment increased the secretion of IL-36α and IL-36γ in AECs ( Fig 2D ) . These data suggested PMs and AECs are likely cellular source of IL-36α and IL-36γ during P . aeruginosa lung infection . Processing and secretion of IL-1β by macrophages in response to P . aeruginosa has been shown to require caspase-1 through activation of the inflammasome , triggered either by flagellin or type-III secretion system [25 , 26] . Recently , we found that IL-36α mRNA expression in AECs is induced by influenza virus through caspase-1 activation dependent [7] . Moreover , other investigators have demonstrated that activation of caspase-1 by Mycobacterium tuberculosis enhanced IL-36γ mRNA in macrophages [27] . We next examined whether activation of caspase-1 contributed to the induction and secretion of IL-36α and IL-36γ by PMs in response to P . aeruginosa . We first confirmed that live P . aeruginosa treated-PMs upregulated the expression of capsae-1p10 , and live , but not heat-killed , bacteria induced the production of IL-1β in CM by PMs ( S1A and S1B Fig ) . Though caspase-1 inhibition attenuated the production of IL-1β by P . aeruginosa-treated PMs , P . aeruginosa-induced IL-36α and IL-36γ mRNA and protein expression were not altered by incubation with a caspase-1 inhibitor ( S1C–S1E Fig ) . These data suggested that unlike influenza and tuberculosis infection , activity of caspase-1 was not involved in the induction and secretion of IL-36α and IL-36γ by PMs in response to P . aeruginosa . Of note , live P . aeruginosa did not increase activity of caspase-3/7 in either primary PMs or AECs ( S2A and S2B Fig ) . To elucidate the functional role of IL-36 receptor ligands during P . aeruginosa pulmonary infection , WT mice , IL-36α deficient ( IL-36α-/- ) , IL-36γ deficient ( IL-36γ-/- ) , and IL-36 receptor ( IL-36R-/- ) mice were challenged intratracheally with 2 × 105 colony forming unit ( CFU ) P . aeruginosa and survival assessed ( Fig 3A ) . Whereas all WT mice were dead within 72 h after bacteria challenge , 80% of IL-36R-/- mice were long term survivors . Importantly , the survival rate of infected IL-36γ-/- mice ( 50% survival ) was also significantly increased compared to WT mice , and was not significantly different with that of IL-36R KO mice . By comparison , the survival rate of infected IL-36α-/- mice was less than 20% and was not statistically different than WT mice . We omitted IL-36α-/- mice from subsequent experiments based on lesser expression of IL-36α relative to L-36γ and lack of survival benefit in infected IL-36α-/- mice . We next examined whether the increased survival rate of P . aeruginosa infected IL-36R-/- and IL-36γ-/- mice was associated with differences in bacterial clearance and dissemination post P . aeruginosa challenge . Bronchoalveolar lavage ( BAL ) and homogenized spleen samples were obtained to determine bacterial loads at 6 h and 24 h after P . aeruginosa infection . At 6 h post bacterial challenge , P . aeruginosa CFU in BAL were similar among WT mice , IL-36R-/- mice and IL-36γ-/- mice . By 24 h , bacteria CFU in BAL were approximately10- and 5-fold lower in IL-36R-/- and IL-36γ-/- mice than in WT mice , respectively ( Fig 3B , left panel ) . Moreover , reduced bacterial dissemination , as assessed by splenic CFU , was observed in IL-36R-/- and IL-36γ-/- mice at 24 h after bacterial administration , as compared to their WT counterparts ( Fig 3B , right panel ) . We next quantified differences in lung inflammatory cell accumulation in WT mice , IL-36R-/- mice and IL-36γ-/- mice during P . aeruginosa infection . No difference in number of total BAL leukocyte and proportion of monocytes/macrophages were observed in uninfected WT mice , IL-36R-/- mice and IL-36γ-/- mice at baseline . Greater than 90% of leukocytes in P . aeruginosa-infected WT mice were neutrophils at 6 h and 24h . Interestingly , we did not find any differences in the number of total cells , neutrophils and monocytes/macrophages among three groups at the selected time points examined during P . aeruginosa infection ( Fig 3C ) . Cytokines and chemokines are important role in host immunity and as mediators of collateral lung injury during experimental P . aeruginosa pulmonary infection [28] . To examine whether IL-36 receptor and IL-36γ genetic deletion altered the production of pro-inflammatory and anti-inflammatory cytokines during P . aeruginosa infection , we measured the levels of TNF-α , IL-6 , and IL-10 in BAL fluid . At 6 h post P . aeruginosa , the levels of these mediators were similar among the three groups . However , at 24 h after P . aeruginosa administration , TNF-α , IL-6 and IL-10 were significantly lower in BAL fluid from IL-36R-/- and IL-36γ-/- mice as compared with WT mice ( Fig 3D ) . We also found significantly lower levels of IL-17 in the BAL fluid of mutant mice at 24 hrs post infection ( Fig 3D ) . IL-17 has been shown to play a critical role in the innate response against extracellular bacterial pathogens , in part through regulating PMN influx and antimicrobial peptide expression . However , there was no difference in BAL PMN accumulation ( Fig 3C ) and mRNA level of β-defensin3 and cathelicidin antimicrobial peptide ( CAMP ) among three groups ( S3 Fig ) . To examine the role of IL-36 ligands in P . aeruginosa induced lung injury , we performed semi-quantitative analysis of lung histology slides from WT mice , IL-36R-/- mice and IL-36γ-/- mice at 10 h and 24 h after P . aeruginosa infection . At 10 h , P . aeruginosa-infected lung of WT mice showed significant histological abnormalities , including alveolar wall edema and inflammatory cells accumulation in the lung interstitium , reflective as a lung injury score of 0 . 71 . In contrast , at this point , P . aeruginosa infected lung histology in IL-36R-/- mice and IL-36γ-/- mice displayed significantly less edema and epithelial cell disruption than that in WT mice ( Fig 4A and 4B ) . At 24 , there was substantial infiltration of neutrophils within the intraalveolar septa and alveolus , as well as alveolar septa edema and proteinaceous debris in the alveolar space , with a similar histological pattern observed in all three groups ( Fig 4C ) . We next assessed the integrity of the alveolar-capillary membrane by measurement of albumin concentration in BAL fluid . At 24 h post infection , albumin levels in BAL fluid were significantly reduced in IL-36R-/- mice and IL-36γ-/- mice compared to WT mice ( Fig 4D ) . PGE2 is known to be an important lipid mediator with a variety of immunosuppressive properties . Previous studies suggest that PGE2 mediates impaired bacterial clearance during P . aeruginosa infection [16 , 18 , 19] . Thus , we first examined the effect of IL-36 receptor ligands on regulating PGE2 production in primary PMs . PMs isolated from WT mice were stimulated with recombinant IL-36α ( 100 ng/mL ) and IL-36γ ( 100 ng/mL ) for 24 h . The expression level of prostaglandin-endoperoxidase synthase 2 ( Ptgs2 ) mRNA was significantly elevated in rIL-36-treated PMs , with rIL-36γ being a more potent inducer of Ptgs2 than rIL-36α ( Fig 5A ) . Induction of Ptgs2 occurred in a dose dependent fashion , as no expression of Ptgs2 mRNA was observed in PMs which was treated with 1 ng/ml or 10 ng/ml concentrations of rIL-36γ . Treatment with rIL-36γ , but not rIL-36α , significantly induced the production of PGE2 in PMs isolated from WT mice . This induction was dependent upon IL-36R signaling , as no induction of PGE2 was observed in PMs isolated from IL-36R-/- mice ( Fig 5B ) , excluding an off target effect of rIL-36γ . We next examined whether autocrine secretion of IL-36 receptor ligands induced by P . aeruginosa stimulation regulated COX-2 expression and PGE2 production in PMs . PMs isolated from WT mice , IL-36R-/- mice and IL-36γ-/- mice were stimulated with P . aeruginosa at a MOI of 10 for 24 h . The production of PGE2 was significantly higher in P . aeruginosa-treated PMs from WT mice than in that from IL-36R-/- mice and IL-36γ-/- mice ( Fig 5C ) . These data suggest that PGE2 synthesis by lung macrophage may be dependent upon autocrine and paracrine IL-36γ secretion in response to P . aeruginosa . We also examined the possibility that AECs could be an important cellular source of PGE2 production in response to rIL-36α ( 100ng/ml ) or rIL-36γ ( 100ng/ml ) . Although the treatment with rIL-36γ induced Ptgs2 mRNA in AECs isolated from WT mice , we did not observe PGE2 production by AECs treated with either rIL-36γ or rIL-36α ( S4A and S4B Fig ) . Next , we examined whether autocrine or paracrine IL-36γ and its receptor; IL-36R contributed to PGE2 synthesis during P . aeruginosa infection in vivo . P . aeruginosa markedly increased the expression of Ptgs2 mRNA in the lung of WT mice at 6 h after infection , whereas we observed significantly lower expression of Ptgs2 mRNA in infected IL-36R-/- mice and IL-36γ-/- mice ( Fig 5D ) . We next examined whether IL-36γ and its receptor contributed to COX-2 protein expression in the lungs during P . aeruginosa infection . Infected lungs of IL-36γ-/- and IL-36R-/- mice showed significantly lower expression of COX-2 compared with infected WT mice by Western blot analysis ( Fig 5E ) . In addition , the production of PGE2 in BAL at 24 h was significantly attenuated in IL-36R-/- mice and IL-36γ-/- mice as compared to infected WT animals ( Fig 5F ) . To confirm the impact of PGE2 on host defense against P . aeruginosa pulmonary infection , we used PGE2-receptor subtype 2 ( EP2 ) receptor deletion mice , as EP2 is a major receptor responsible for the immunosuppressive and anti-inflammatory properties of PGE2 [29] . Previous study demonstrated that the profile of EP receptor in PMs is EP2>EP1>EP4>EP3 and activated macrophages increased the expression of EP2 and decreased EP4 expression [18] . WT mice and EP2-/- mice were treated intratracheally with 2 . 0 × 105 CFU P . aeruginosa , and then quantitated bacterial burden in BAL and spleen , the production of pro-inflammatory cytokines and albumin concentration in BAL at 24 h post infection ( Fig 6A–6C ) . Bacterial loads in BAL and spleen in EP2-/- mice were 2 . 7- and 5 . 8-fold lower compared to WT mice , respectively ( Fig 6A ) . However , we did not find differences in levels of TNF-α and IL-6 ( Fig 6B ) or albumin concentration ( Fig 6C ) in BAL between infected WT mice and EP2-/- mice . Finally , to confirm contribution from the COX-2/PGE2 pathway in host defense against P . aeruginosa pneumonia , we administrated the COX-2-specific inhibitor NS-398 ( 10μM ) to WT mice i . p . 1 h before bacterial challenge . As shown Fig 6D and 6E , treatment with NS-398 result in a trend toward to reduce the bacterial CFU in BAL and significantly reduced the dissemination in P . aeruginosa infected mice . Similar to observations in EP2-/- mice , we observed no difference in BAL albumin concentration between treatment with NS-398 and vehicle control . These data suggest that COX2/PGE2/EP2 signaling in this model can promote impaired antimicrobial immunity but does not appear to regulate inflammatory cytokine production or lung injury responses . We next examined whether IL-36γ directly regulated P . aeruginosa phagocytosis and bacterial killing by PMs , PMs were isolated from WT mice and IL-36R-/- mice , then treated with rIL-36γ ( 100ng/ml ) for 18h , then cells incubated with FITC-labeled heat-killed P . aeruginosa . PMs isolated from WT and IL-36R-/- mice showed similar phagocytosis of FITC-labeled bacteria . Additionally , treatment with rIL-36γ did not alter the phagocytic ability of PMs isolated from WT mice ( Fig 7A ) . For bacterial killing assays , PMs were incubated with live PA , washed , and viable intracellular CFU determined by standard culture techniques . The number of viable intracellular bacteria in PMs isolated from WT mice and IL-36R-/- mice was similar at 30 min and 120 min after P . aeruginosa inoculation . Interestingly , IL-36γ treated PMs from WT mice displayed impaired bactericidal activity , as evident by a significantly higher number of viable bacteria at 120 min post bacterial inoculation than non-treated PMs ( Fig 7B ) . This effect was dependent on specific IL-36 receptor signaling , as rIL-36γ treatment did not alter bacterial killing in PMs isolated from IL-36R-/- mice . To examine whether PGE2 might be responsible for IL-36γ- mediated impairment in microbicidal activity , we isolated PMs from WT mice and mice lacking the EP2 receptor . Killing of intracellular bacteria by PMs isolated from EP2-/- mice was similar to PMs isolated from WT mice . However , Treatment of WT PM with rIL-36γ resulted in decreased microbicidal activity , whereas incubation with rIL-36γ did not inhibit microbicidal activity in PMs lacking EP2 ( Fig 7C ) . In addition , we examined whether IL-36γ-induced COX-2 mediated impaired bacterial killing by PMs . Treatment with NS-398 enhanced bacterial killing activity in IL-36γ treated PMs ( S5E Fig ) . Thus , IL-36γ-induced impairment in PM microbicidal activity requires COX-2/PGE2/EP2 signaling . Finally , to determine whether observations in P . aeruginosa murine pneumonia model were of potential clinical relevance , we measured levels of IL-36α and IL-36γ in plasma and BAL fluid of patients with acute respiratory distress syndrome ( ARDS ) caused by P . aeruginosa within 7 days from the onset of ARDS . We found that IL-36γ were significantly elevated in both plasma and BAL samples of patients with P . aeruginosa induced ARDS patients , as compared to healthy subjects ( Fig 8A and 8B ) . By comparison , a trend toward elevated IL-36α levels was found in plasma but not BAL of these patients , as compared to healthy subjects .
In Gram-negative bacterial pneumonia , the innate immune system , including neutrophils and macrophages , plays a necessary role in the rapid clearance of pathogens from the lung . However , a vigorous inflammatory response against microbes , particular pathogens that express a broad armament of virulence factors , can promote collateral damage to lung tissue resulting in acute lung injury [30] . In this study , we demonstrate that IL-36γ released from PMs , AECs and likely other cellular source is not only dispensable for bacterial clearance , but can actually inhibit antimicrobial immunity , an effect that is at least partially PGE2-dependent . Moreover , IL-36 ligand can promote lung injury by a mechanism that is likely independent of PGE2 . Stimulation with bacteria and virus induces IL-36 cytokines from epithelial cells , macrophages and monocytes in the lungs [6 , 7 , 27] . We have recently shown that influenza virus stimulates production of IL-36α by AECs , and IL-36 receptor deficient mice prevent from influenza virus-induced lung injury [7] . However , current knowledge of the functional role of IL-36 receptor ligands in bacterial infection is limited . We found significantly increased level of IL-36α and IL-36γ in the lungs of P . aeruginosa infected mice ( Fig 1 ) , and P . aeruginosa-treated PMs and AECs expressed IL-36α and IL-36γ ( Fig 2 ) . Importantly , the functional phenotype IL-36γ-/- mice during infection resembled that of IL-36R-/- mice , whereas survival of IL-36α-/- mice paralleled that of infected WT mice ( Fig 3A ) . We cannot exclude a minor contribution of IL-36α to the phenotype observed , but IL-36γ appears to be predominant IL-36R agonist responsible for the observed effects . The quantity of IL-36α in P . aeruginosa infected lungs was considerably less than that of IL-36γ production . In vitro , treatment with rIL-36γ , but not IL-36α significantly induced COX-2 mRNA expression and the production of PGE2 in PMs . Findings in the murine model are in keeping with our observations in patients with Pseudomonas pneumonia and ARDS , as we observed significantly higher levels of IL-36γ , but not IL-36α , in plasma and BAL as compared with healthy control subjects ( Fig 8 ) . Collectively , these data suggest that IL-36γ is the predominant IL-36R ligand involved in the pathogenesis of P . aeruginosa induced severe pneumonia . Of note , all patients with Pseudomonas pneumonia also had ARDS , so we were unable to distinguish the influence of pseudomonal infection alone from that of the immune landscape of acute lung injury . A somewhat unanticipated finding was the observation that IL-36R-/- mice and IL-36γ-/- mice displayed decreased bacterial load in the alveolar space and less dissemination during P . aeruginosa infection ( Fig 3B ) . This is unlikely to be a direct effect of IL-36 agonists on neutrophil accumulation and activation , as we observed no differences in PMN numbers and mouse neutrophils do not express IL-36 receptor [31 , 32] . Furthermore , we did not observe differences in phagocytosis and bacterial killing in neutrophils isolated from WT mice , IL-36R-/- mice and IL-36γ-/- mice during P . aeruginosa infection in vitro ( S5A and S5B Fig ) . Also , there were no differences in phagocytic and microbicidal activity of PM isolated from WT mice , IL-36R-/- mice and IL-36γ-/- mice ( S5C and S5D Fig and Fig 7A and 7B ) . These data implicate other host derived mediators induced by IL-36γ that contribute to impaired bacterial clearance . It is worth noting that there was a more impressive reduction in bacterial dissemination ( as reflected by splenic CFU ) in IL-36R-/- mice and IL-36γ-/- mice when compared to WT animals than reductions in BAL CFU observed in these animals , which is more modest . We believe this is attributable to less lung injury and disruption of the alveolar-capillary membrane in the mutant mice post P . aeruginosa challenge . Our findings implicate PGE2 as a relevant contributor to IL-36-mediated suppression of anti-pseudomonal immunity . PGE2 is a well-recognized lipid regulator of inflammatory and immune responses during acute and chronic infections . IL-1β , a family member of IL-36 cytokines , is known to induce COX-2 expression and production of PGE2 in macrophages [33 , 34] , and the induction of macrophage-derived PGE2 in response to Mycobacterium tuberculosis is dependent upon IL-1 receptor ligands [35] . However , interactions between IL-36 receptor ligands and PGE2 expression have not previously been described . We found that rIL-36γ , but not rIL-36α , dose-dependently induced Ptgs2/COX-2 mRNA expression and PGE2 production in PMs , and that COX-2 expression and PGE2 production by PMs isolated from IL-36R-/- mice and IL-36γ-/- mice in vitro were impaired relative to PMs isolated from WT mice ( Fig 5 ) . Vigne et . al . demonstrated that rIL-36β the induction of IL-6 , CXCL-1 , CCL1 and IL-23p19 mRNA by rIL-36β was 1 . 5–2 . 0 fold greater than that induced by rIL-36α and IL-36β in bone marrow derived dendritic cells [31] in vitro . Other investigator have shown that rIL-36γ , but not rIL-36α significantly induced TNF-α , and only IL-36β could induce β-defensin ( HBD ) -2 , HBD-3 and CAMP in human keratinocytes [21] . Collectively , these studies suggest that response to IL-36 cytokines is depends on both the specific IL-36 ligand and the cells which expresses IL-36 receptor . In addition , the in vivo expression of COX-2 and production of PGE2 in the lungs of P . aeruginosa-infected IL-36R-/- mice and IL-36γ-/- mice was mitigated as compared to infected WT animals ( Fig 6B and 6C ) . Importantly , the defect in PGE2 production was similar in IL-36R-/- mice and IL-36γ-/- mice . Taken together , endogenously produced IL-36γ and its receptor signal contribute to the induction of PGE2 during P . aeruginosa infection . Previous studies have demonstrated a relevant immunomodulatory influence of prostaglandins on antimicrobial function of phagocytes , including PMN and mononuclear phagocytes . For instance , activation and aggregation of neutrophils is inhibited after exogenous treatment with PGE2 in vitro [36 , 37] . In addition , PGE2 impaired the ability of PMN to kill P . aeruginosa [18] . In other murine P . aeruginosa infection models , PGE2 has been shown to impair both internalization and killing of ingested bacteria by macrophages [18 , 17] . In our study , we did not find differences in phagocytic properties ( Fig 7A ) and initial intracellular bacterial loads in PMs ( Fig 7B ) between IL-36γ and vehicle treated PMs in vitro . In addition , no significant difference in BAL bacterial CFU was observed among infected WT mice , IL-36R-/- mice and IL-36γ-/- mice at early period post infection ( 6 hrs , Fig 3B ) in vivo . These data indicates that IL-36γ does not contribute to initial bacteria uptake by resident leukocytes . Reactive oxygen species and nitric oxide ( NO ) are important mediators of macrophages bactericidal activity in P . aeruginosa infection [38 , 39] . PGE2 has been shown to suppress NO synthesis in murine macrophages [15] , and microbicidal activity in phagocytic cells [38 , 39] . We found impaired bacterial killing in PMs treated with IL-36γ , and this effect was mitigated in PMs isolated from EP2 receptor deficient mice ( Fig 7B and 7C ) . In addition , EP2-/- mice showed decreased bacterial loads in BAL and attenuated dissemination in spleen compared with WT mice during P . aeruginosa infection ( Fig 6C ) , findings that mirrored that observed in both IL-36γ-/- and IL-36R deficient mice . While these observations point to PGE2 as a major mediator of impaired antimicrobial immunity in IL-36γ/IL-36R mutant mice , there are likely other mediators involved . One such candidate is IL-10 , which can suppressive antimicrobial responses and the in-vivo production of IL-10 was reduced in both infected IL-36R-/- mice and IL-36γ-/- mice . However , inhibition of IL-10 bioactivity by neutralizing antibody administration did not alter bacterial clearance in our model ( S6 Fig ) , suggest that IL-10 was not responsible for the effects observed . The finding of improved P . aeruginosa clearance and reduced dissemination in IL-36γ-/- mice differs from observations we have recently made in another murine Gram-negative bacterial pneumonia model ( Klebsiella pneumonia ) [40] . In this model , we observed that IL-36γ-/- mice were more susceptible to Klebsiella pneumoniae challenge due to impaired expression of type 1 and IL-17 cytokines . There may be several reasons for these disparate findings . First , K . pneumoniae is a heavily encapsulated organism that is much more invasive than P . aeruginosa , and a vigorous type 1 and IL-17 innate response is required for clearance from the airspace . By comparison , challenge with P . aeruginosa , especially cytotoxic strains such as ATCC 19660 , results in marked and deleterious inflammation and injury , and invasion only occurs with high inoculum of organisms . Indeed , in pneumonia caused by Klebsiella pneumonia murine model , pro-inflammatory cytokines such as TNF-α , IL-1 , and IL-17 [41 , 42] are required for bacteria clearance from the lungs , whereas the anti-inflammatory cytokine IL-10 impairs host defense in this infection model [43] . Conversely , the effect of these pro-inflammatory and anti-inflammatory cytokine in the host defense against P . aeruginosa is opposite to that observed in K . pneumoniae infection [44–46] . Moreover , the peak of lung IL-36γ expression after bacteria challenge is earlier post P . aeruginosa administration as compared to K . pneumoniae administration , which may be partially attributable to the much higher inoculum used in the Pseudomonas model [40] , We speculate that early and marked cytokine storm induced by cytotoxic strains of P . aeruginosa promotes deleterious lung injury , Moreover , PGE2 is highly induced and plays an important immunoregulatory role in Pseudomonas lung infection , whereas this has not been as convincingly shown in more progressive and invasive infections such as K . pneumoniae . Lung injury was mitigated in IL-36R-/- mice and IL-36γ-/- mice during P . aeruginosa infection , as indicated by reduced lung injury scores and lower BAL albumin concentration ( Fig 4E ) . The P . aeruginosa strain ( ATCC 19660 ) we used in the present study expresses the Type III secretion system ( T3SS ) , which deliver virulence factors to the cytosol of host cells , and the T3SS in P . aeruginosa promotes lung injury through disruption of alveolar-epithelial barrier [47 , 48] . The mechanism ( s ) by which IL-36 cytokines exacerbate lung injury in P . aeruginosa pneumonia have not been completely defined . This effect is unlikely to be mediated by PGE2/EP2 signaling , as no differences in BAL albumin levels were noted between EP2 deficient mice or mice treated with NS-398 as compared to control mice post P . aeruginosa challenge despite differences in lung bacterial burden . Also , we did not find differences in inflammatory leukocyte influx at 6 and 24 h post bacterial administration . We did note reductions in both IL-6 and IL-17 , and to a lesser extent TNFα in the BAL fluid of infected IL-36R-/- mice and IL-36γ-/- mice at 24 h post infection , as compared to WT mice . In patients with community-acquired pneumonia , the levels of IL-6 in BAL and plasma are positively correlated with the severity of disease [49 , 50] . Moreover , IL-6 deficient mice have been shown to be protected from lung injury and mortality during P . aeruginosa pulmonary infection [51] . Mechanistically , IL-6 can induces signal transducers and activator of transcription ( STAT ) 3 activation , and excessive lung STAT3 activation in P . aeruginosa lung infection has been shown to result in more severe lung injury and increased mortality [52] . IL-36 receptor ligands directly induce the production of IL-6 in PMs , AECs [7] and dendritic cells [31] . These data suggested that IL-6 induced by IL-36γ may be associated with lung injury during P . aeruginosa infection . We also observed reductions in IL-17 in infected mutant mice , and IL-17 has been shown to mediate injury responses in certain pulmonary infections such as pneumonia caused by influenza virus [53] . However , we did not observe defects in BAL PMN accumulation ( Fig 3C ) or antimicrobial peptide mRNA expression in IL-36R or IL-36γ deficient mice ( S3 Fig ) , calling into question the physiological significance of reduced IL-17 levels . In conclusion , this study identifies IL-36γ , released from PMs and AECs and likely other lung cells , as a mediator of impaired lung host immune response and lung injury during P . aeruginosa infection . Our findings provide fundamental insights into the pathophysiology of P . aeruginosa induced pneumonia , insights which may have important therapeutic implications .
Specific pathogen-free age- and sex-matched C57BL/6 mice were purchased from The Jackson Laboratory ( Bar Harbor , ME , USA ) . IL-36R-/- mice on a C57BL/6 background were provided by Jennifer Towne from Amgen ( Thousand Oaks , CA , USA ) [14] . A colony of IL-36α-/- mice bred on a C57BL/6 background was established and provided from RIKEN BRC ( Tsukuba , Japan ) . A colony of IL-36γ-/- mice bred on a C57BL/6 background was established at the University of Michigan ( Ann Arbor , MI , USA ) [6] . EP-2-/- mice on a C57BL/6 background were kindly given to us by Marc Peters-Golden at the University of Michigan [54] . All mice were housed in specific pathogen-free conditions within the University of Michigan Animal Care Facility . Flagellated P . aeruginosa strain ATCC 19660 ( American Type Culture Collection , Manassas , VA , USA ) was used for all of experiments . Bacteria was grown overnight in Difco nutrient broth ( BD Biosciences , Franklin Lakes , NJ , USA ) at 37°C with constant shaking . Bacteria concentrations were determined by measuring the amount of absorbance at 600nm and compared to a predetermined standard curve based on known colony-forming unit ( CFU ) values . To prepare heat killed P . aeruginosa , bacteria was incubated at 65°C for 1h . For FITC labeling , heat-killed P . aeruginosa was resuspended at 109–1010 CFU/mL in 0 . 1M NaHCO3 ( pH 9 . 0 ) . A total of 0 . 2 mg/ml FITC ( Invitrogen , Carlsbad , CA , USA ) in DMSO was added to heat-killed bacteria and incubate in the dark for 1 h on a rocker at room temperature . Following FITC labeling , bacteria was washed three times and responded in 1ml sterile PBS . Aliquots were prepared and store at -80°C . For intratracheally administration , mice were anesthetized by intraperitoneal injection of ketamine and xylazine and then infected 50μl of 2 × 105 CFU by insertion of 24-gauge intravenous catheter into the trachea . BAL was performed as described previously [55] . Mice were euthanized by CO2 inhalation . The trachea was exposed and cannulated with 22 G intravenous catheter . BAL was performed with 3 mL PBS containing 5mM EDTA ( tree aliquot 1mL of PBS ) , and then pulmonary circulation was rinsed by 1ml PBS . Lungs were harvested for RNA extraction , immediately snap-frozen in liquid nitrogen . After a collection of leukocytes in BAL fluids , cytospin ( 113 g × 5 min ) preparations were made from each sample and stained with modified Wright stain . Differential cell counts of neutrophils and monocytes and macrophages were obtained for at least 400 cells counts in each sample at a magnification of ×1000 . Samples of BAL fluid and homogenized spleen in PBS were serially diluted 10 fold in PBS . 10 μl of each samples were plated on a nutrient agar . Bacterial colonies were counted after the plates were incubated at 37°C for 18 h . Lungs and trachea were removed from euthanized animals and inflated at 20cm H2O with 4% paraformaldehyde through trachea , and fixed for paraffin embedding . All lungs were sectioned and stained with haematoxylin and eosin ( H&E ) . Quantitative analysis of tissue injury was measured using the lung injury scoring system as described [56] . Lung injury scoring system parameters include neutrophils in the alveolar space ( A ) , neutrophils in the interstitial space ( B ) , hyaline membranes ( C ) , proteinaceous debris filling the airspaces ( D ) and alveolar septal thickening ( E ) . At least 20 random regions were scored 0–2 independently at a magnification of ×400 in a blinded fashion . The final lung injury score per each lungs was calculated as below; score = [ ( 20 × A ) + ( 14 × B ) + ( 7 × C ) + ( 7 × D ) + ( 2 × E ) ] / ( number of fields × 100 ) . Murine pulmonary macrophages ( PMs ) and type II alveolar epithelial cells ( AECs ) were isolated using the method described previously [57 , 58] . Briefly , pulmonary macrophages ( consisting of both alveolar and interstitial macrophages ) were isolated from dispersed lung digest cells by adherence purification as previously described [57] . For the isolation of murine AECs , the pulmonary vasculature was perfused . The lungs were filled via the trachea with 1 . 5 ml dispase ( Worthington , Lakewood , NJ . USA ) , then 1 . 5 ml of low-melting point agarose and finally placed in ice cold PBS . The lungs were submerged in dispase for 45 min at 24°C before the lung tissue was teased from the airways and minced in DMEM with 0 . 01% DNase . After swirling for 15 min , followed by passage through a series of nylon filters , the cell suspension was collected by centrifugation and incubated with biotinylated Abs ( anti-CD32 and anti-CD45; BD Pharmingen , San Diego , CA , USA ) . After incubation with streptavidin-coated magnetic particles , myeloid cells were removed using a magnetic tube separator . Mesenchymal cells were removed by overnight adherence in a Petri dish and the resulting non-adherent cells were plated on plastic dishes coated with fibronectin . Previous work has shown that the day 3 time point has >90% pure AECs [58] . These cells were treated with live or heat killed P . aeruginosa at a MOI of 10 , LPS ( 1μg/ml ) ( Sigma-Aldrich ) , and recombinant IL-36α ( 100ng/mL ) and IL-36γ ( 100ng/mL ) ( R&D Systems Minneapolis , MN , USA ) . The ability of PMs to phagocytosis bacteria was examined using FITC-labeled P . aeruginosa . PMs isolated from WT mice , IL-36R-/- mice and EP2-/- mice were plated at 1 × 106 cells/well and cultured overnight with or without rIL-36γ ( 100ng/mL ) on 24-well plate . The following day , wells were washed with antibiotics free CM , and PMs were incubated with FITC-labeling or non-labeling heat-killed P . aeruginosa at a MOI 100 . Two hours later , cells were collected by cell scraper and stained with PerCP-Cy5 . 5-labeling CD45 ( BD Pharmigen , San Jose , CA , USA ) and PE-labeling F4/80 ( BD Pharmigen ) . Isotype controls were used for all the samples . PMs phagocytosis of FITC-labeled bacteria was analyzed by Attune Acoustic Focusing Cytometer ( Thermos Scientific-Applied Biosystems , Foster City , CA , USA ) . Bacterial killing assay was assessed using a modification of protocol previously reported [59] . PMs were seeded on 24-well plate at 1 × 106 cells/well and cultured overnight with or without rIL-36γ ( 100ng/ml ) . Following day , PMs were infected with live P . aeruginosa at a MOI 100 . At 30 min after incubation , infected PMs were washed with gentamicin at 100μg/ml solution twice to remove extracellular bacteria , and then cells were lysed to obtain initial CFU , or incubated further at 37°C for more 90 min . Cells were subsequently lysed by 0 . 1% Triton X-100 , followed by serial plating for bacterial CFU quantification . Murine IL-36α and IL-36γ secreted in BAL and CM were measured by previously reported sandwich ELISA method [6] . For human IL-36α and IL-36γ ELISA generation , human recombinant IL-36α and IL-36γ and human anti-IL-36α and anti-IL-36γ polyclonal antibodies were purchased from R&D Systems . Other cytokines/chemokines ( TNF-α , IL-6 , IL-17 and IL-10; R&D systems ) and albumin ( Albumin Quantification Kit: Bethyl Laboratories , Montgomery , TX , USA ) were quantified using a modified double-ligand method as described . The production of PGE2 were determined using an ELISA Kit according to the manufacture’s protocol ( Cayman , Ann Arbor , MI , USA ) . RNA was isolated and real time quantitative RT-PCR was performed by AB Step One plus Real-Time PCR System ( Thermos Scientific-Applied Biosystems ) . Predesigned primer and probes of targeted molecules and β-actin as a control were purchased from Integrated DNA Technologies ( Coralville , IA , USA ) . Quantification of β-actin and target genes in each sample set was performed by the standard curve method . Cells were digested by RIPA buffer ( Sigma-Aldrich ) plus protease inhibitors and gels were subjected to electrophoresis as previously described [58] . Membranes were incubated with primary anti-COX-2 antibody ( Cayman; diluted 1:100 ) or β-actin ( Sigma-Aldrich; diluted 1:20 , 000 ) , blots were incubated with a secondary antibody linked to HRP and the signals were developed with an ECL ( SuperSignal West Pico Substrate , Pierce Biotechnology , Rockford , IL , USA ) . Patients with ARDS that were enrolled in the Acute Lung Injury Specialized Center of Clinically Oriented Research ( SCCOR ) randomized trial of granulocyte-macrophages colony stimulating factor administration in ARDS conducted at the University of Michigan between Jan 2004 and October 2007 were studied [60] . We identified patients with P . aeruginosa induced ARDS who obtained P . aeruginosa from sterile culture sites including blood or BAL samples , and no other putative pathogens identified . Sixteen patients were matched for these criteria , and 16 plasma and 10 BAL samples were obtained from these patients within seven days from the onset of ARDS . Six plasma and nine BAL samples of healthy subjects were used as control . Animal studies were reviewed and approved by the University Committee on Use and Care of Animals at the University of Michigan in accordance with guidelines of the Care and Use of Laboratory Animals of the National Institutes of Health ( protocol #PRO00006295 ) . Experiments using human samples were approved by the University of Michigan Institutional Review Board ( IRB#2003–0430 and IRB#2003–0829 ) and conducted in accordance with the principles expressed in the Declaration of Helsinki . Written informed consent was obtained from participants . If the patients with ARDS received mechanical ventilation under the sedation , and they were not able to make their decision by themselves , we obtained written informed consent from their legal proxy for medical decision making before study inclusion . IRB approved a legally authorized representative to sign proxy informed consent . All patients enrolled in this study were over the age of 18 . Descriptive statics , such as means and standard deviations , were collected . The difference in survival rates was evaluated by the log rank test ( Mantel-Cox ) . Two sets of values were evaluated by the Student’s t-test , and more than three sets of value were evaluated by ANOVA , followed by the Turkey’s multiple comparison test . Data analysis was conducted using Graphpad prism 6 ( GraphPad Software , La Jolla , CA , USA ) . A P value of <0 . 05 was considered satirically significance . | Pneumonia caused by Pseudomonas aeruginosa is a serious infection resulting in significant lung injury and mortality in susceptible hosts . The first line of defense in P . aeruginosa lung infection are neutrophils and macrophages , which play a pivotal role in the rapid clearance of pathogens from the lung . However , excessive innate responses against microbial pathogens can promote collateral damage to lung tissue , culminating in acute lung injury . Here , we demonstrated the role of IL-36 cytokines in modulating the innate immune response to P . aeruginosa pulmonary infection . The elaboration of interleukin ( IL ) -36γ in the alveolar space was observed not only in mice during P . aeruginosa pulmonary infection but also patients with pneumonia due to P . aeruginosa . In murine P . aeruginosa lung infection , deletion of IL-36γ or its receptor resulted in improved bacterial clearance associated with reduced prostaglandin E2 production , and attenuated lung injury independent of changes in leukocyte influx . Taken together , blockage of IL-36γ and its receptor signal may represent a viable immunomodulatory therapeutic approach in cytotoxic pseudomonas respiratory infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"immune",
"cells",
"pathogens",
"immunology",
"microbiology",
"pulmonology",
"pseudomonas",
"aeruginosa",
"animal",
"models",
"pneumonia",
"bacterial",
"diseases",
"developmental",
"biology",
"model",
"organisms",
"experimental",
"organism",
"systems",
"molecular",
"development",
"pseudomonas",
"infections",
"bacteria",
"neutrophils",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"pseudomonas",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"mouse",
"models",
"immune",
"system",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2017 | Interleukin-36γ and IL-36 receptor signaling mediate impaired host immunity and lung injury in cytotoxic Pseudomonas aeruginosa pulmonary infection: Role of prostaglandin E2 |
Four dengue virus serotypes ( DENV1-4 ) circulate globally , causing more human illness than any other arthropod-borne virus . Dengue can present as a range of clinical manifestations from undifferentiated fever to Dengue Fever to severe , life-threatening syndromes . However , most DENV infections are inapparent . Yet , little is known about determinants of inapparent versus symptomatic DENV infection outcome . Here , we analyzed over 2 , 000 DENV infections from 2004 to 2011 in a prospective pediatric cohort study in Managua , Nicaragua . Symptomatic cases were captured at the study health center , and paired healthy annual samples were examined on a yearly basis using serological methods to identify inapparent DENV infections . Overall , inapparent and symptomatic DENV infections were equally distributed by sex . The mean age of infection was 1 . 2 years higher for symptomatic DENV infections as compared to inapparent infections . Although inapparent versus symptomatic outcome did not differ by infection number ( first , second or third/post-second DENV infections ) , substantial variation in the proportion of symptomatic DENV infections among all DENV infections was observed across study years . In participants with repeat DENV infections , the time interval between a first inapparent DENV infection and a second inapparent infection was significantly shorter than the interval between a first inapparent and a second symptomatic infection . This difference was not observed in subsequent infections . This result was confirmed using two different serological techniques that measure total anti-DENV antibodies and serotype-specific neutralizing antibodies , respectively . Taken together , these findings show that , in this study , age , study year and time interval between consecutive DENV infections influence inapparent versus symptomatic infection outcome , while sex and infection number had no significant effect . Moreover , these results suggest that the window of cross-protection induced by a first infection with DENV against a second symptomatic infection is approximately 2 years . These findings are important for modeling dengue epidemics and development of vaccines .
Dengue is a major health problem globally , with more than 40% of the world's population at risk and over a hundred countries affected by epidemics [1] . In the past 50 years , the incidence of dengue has increased considerably , affecting tens of millions of people annually . Dengue is caused by an enveloped , positive-sense RNA virus in the genus Flavivirus of the Flaviviridae family , which is transmitted by mosquitoes of the Aedes genus . There are four serotypes of dengue virus ( DENV ) : DENV-1 , DENV-2 , DENV-3 and DENV-4 . Infection with DENV can be subclinical ( inapparent infection ) or cause a variety of clinical manifestations ranging from undifferentiated illness and Dengue Fever ( DF ) to severe life-threatening syndromes Dengue Hemorrhagic Fever ( DHF ) and Dengue Shock Syndrome ( DSS ) [2] . Very little is known about the determinants of inapparent versus symptomatic DENV infection outcome . By definition , inapparent infections are not detected in routine surveillance and can only be captured in the context of prospective cohort or index cluster studies . In a cohort study in Thailand , study year , total DENV infection incidence in the current and previous year , circulating DENV serotype and the number of circulating serotypes were identified as factors influencing inapparent versus symptomatic infection outcome [3] , [4] . Analysis of infection outcome is further complicated by immune responses to multiple infections with different DENV serotypes , which can be either protective or pathogenic . Early experimental studies in DENV-naïve healthy volunteers showed that infection with one DENV serotype confers immunity to that particular serotype for up to 18 months [5] . In fact , this protection is thought to be life-long . On the other hand , infection with one serotype only conferred short-term ( <2 months ) complete protection against heterologous infection with a different serotype [5] . In Sabin's studies , heterologous protection waned over a period of several months . Heterologous protection after a short interval but not after a longer period of time was also observed in rhesus monkeys , depending on the serotype sequence [6] . In contrast , secondary heterologous infection is well documented as the single most important risk factor for severe dengue [7]–[11] . Epidemiological data from dengue epidemics in Cuba also suggest that longer time intervals between infections might increase disease severity [11] . In 1977 , DENV-1 caused the first dengue epidemic in the country . This was followed by two DENV-2 epidemics caused by similar strains in 1981 and 1997 , respectively [12] , [13] . Interestingly , death rates were significantly higher in 1997 compared to 1981 [12] . Altogether , these observations highlight the intricate interplay between host immunity and repeat DENV infections and suggest that the time between two consecutive infections is an important factor in infection outcome . Few studies have compared inapparent versus symptomatic outcome in primary and secondary DENV infections . In one of the first prospective dengue cohort studies in Bangkok , Thailand [9] , and in a multinational index cluster study with four sites in South-East Asia and Latin America [14] , the inapparent-to-symptomatic ratio was similar in primary and secondary infections . We also previously reported similar ratios in primary and secondary DENV infections in Managua , Nicaragua [15] . However , an index cluster study conducted in Kamphaeng Phet , Thailand , found very few symptomatic dengue cases among primary infections when compared to secondary infections , albeit the overall number of infections reported in the study was limited [16] . Even less is known about the impact of second , third or fourth DENV infections ( collectively referred to as “secondary infections” ) on inapparent versus symptomatic outcome . In fact , few reports exist in the literature of third and fourth DENV infections [17] , [18] . In a hospital-based retrospective study , third and fourth DENV infections were estimated to present a lower risk of hospital admission [19] . However , once hospitalized , the risk of DHF/DSS in third and fourth DENV infections was not different from that in second DENV infections [19] . In Nicaragua , the first dengue epidemic was reported in 1985 and caused by DENV-1 and DENV-2 [20] . Several DENV-1 , 2 and 4 outbreaks occurred in the early 1990's , followed by a large DENV-3 epidemic in 1994–5 [21] . Since then , all four serotypes circulate , but in contrast to hyperendemic areas , one serotype is dominant each season [22]–[24] . The dengue season starts after the first rains , with most cases occurring from August to January [15] . However , some cases are detected throughout the year . In 2004 , we established the community-based , prospective Pediatric Dengue Cohort Study ( PDCS ) in Managua , Nicaragua [25] . Here , we analyzed serological data from all cohort participants , as well as neutralizing antibody titers in a subset of children who had experienced repeat DENV infections , using 8 annual healthy blood sample collections . We combined these results with data about dengue cases in the PDCS from 7 dengue seasons to investigate the determinants of inapparent versus symptomatic DENV infection outcome . In particular , we evaluated the impact of factors that can only be analyzed in the context of long-term cohort studies such as infection number and the time interval between infections in children with documented repeat DENV infections .
This study was approved by the Institutional Review Boards of the Nicaraguan Ministry of Health and the University of California , Berkeley . Parents or legal guardians of all subjects provided written informed consent , and subjects 6 years of age and older provided assent . In August of 2004 , a community-based pediatric dengue cohort study was established in District II of the capital city of Managua , a low-to-middle income area with a population of approximately 62 , 500 [25] . This ongoing study is based at the municipal Health Center Sócrates Flores Vivas ( HCSFV ) , which is the principal source of primary health care for the district's population . Initially , participants aged 2–9 years were recruited through house-to-house visits; over time , the age range of the study was extended to 2 to 14 years of age . Each year , additional children were enrolled to maintain the age structure of the cohort [25] . Participants were encouraged to seek medical care for all illnesses through study physicians and in particular , to present early in case of a febrile episode . Cohort participants were followed closely for all illnesses , and study physicians classified participants into febrile illnesses that met the WHO dengue case definition ( category A ) [2] , undifferentiated fever ( category B ) , fever with an apparent focus other than dengue ( category C ) , and non-febrile episode ( category D ) . Children who met WHO criteria for suspected dengue ( category A ) as well as those with undifferentiated fever ( category B ) were evaluated for acute DENV infection [15] , [25] . The cohort was sized such that even in years of relatively low DENV transmission , a minimum number of symptomatic cases would be identified . A suspected dengue case was considered a symptomatic DENV infection when 1 ) DENV RNA was detected by reverse-transcriptase polymerase chain reaction ( RT-PCR ) [26] , [27] , 2 ) DENV was isolated [26] , 3 ) seroconversion was observed in paired acute and convalescent phase sera by IgM capture ELISA [26] , [28] , or 4 ) seroconversion and/or a ≥4-fold increase in total DENV-specific antibody titer in paired acute and convalescent sera was observed by Inhibition ELISA [29] , [30] . Inapparent DENV infections were identified through serological testing of paired annual blood draws from healthy subjects [15] , [25] . Participants whose paired annual samples demonstrated seroconversion or a 4-fold or greater increase in total DENV-specific antibody titer by Inhibition ELISA , but who had not experienced a documented febrile episode associated with acute DENV infection , were considered to have experienced an inapparent DENV infection [15] , [25] . To evaluate the effectiveness of capture of febrile cases , yearly participant surveys were conducted ( Table S1 ) . Overall , surveys showed that only 1 . 9% of the participants reported having a fever and attending a different healthcare provider and 2 . 3% reported not attending any medical provider . Both symptomatic and inapparent DENV infections were assigned a dengue season whose limits were defined by the healthy annual blood collection . As a specific date cannot be assigned to inapparent DENV infections , since by definition the infection is inapparent and thus not reported to the study health center , the inapparent infection date was assumed to be October 1st , during the peak of the corresponding season . For consistency , the same procedure was followed for symptomatic infections . Raji-DC-SIGN cells ( kind gift from B . Doranz , Integral Molecular , Philadelphia , PA ) were used for all neutralization experiments . Cells were grown at 37°C at 5% CO2 in RPMI medium supplemented with 10% ( v/v ) Fetal Bovine Serum ( FBS ) , 1% ( v/v ) penicillin-streptomycin , and 0 . 1 M HEPES ( RPMI complete medium ) . DC-SIGN ( CD209 ) expression was quantified by flow cytometry using a monoclonal antibody ( PerCP-Cy5 . 5 Mouse Anti-Human CD209 , BD Biosciences ) , and cell lots were used only if >95% of the cells were positive for DC-SIGN . DENV Reporter Viral Particles ( RVP; DENV-1 , Western Pacific 74; DENV-2 , S16803; DENV-3 , CH53489; DENV-4 , TVP360 ) containing a GFP reporter RNA were produced by Integral Molecular as previously described [31] , [32] . RVPs were stored at −80°C , and for experiments , were thawed rapidly in a water bath and kept on ice before use . For each RVP lot , the optimal working dilution was determined . Briefly , RVPs were serially diluted 2-fold in RPMI complete medium adjusted to pH 8 . 0 with 5 M NaOH . Infection was carried out in a 96-well plate by mixing , in each well , 50 µl of diluted RVPs with 40 , 000 Raji DC-SIGN cells in a total volume of 100 µl complete RPMI media . The cells were then incubated at 37°C in 5% CO2 for 48 hours and subsequently fixed in 2% paraformaldehyde . The percentage of infected , GFP-expressing cells was determined by flow cytometry ( Becton-Dickinson LSRII or Beckman Coulter Epics XL-MCL ) using FlowJo version 7 . 2 . 5 ( TreeStar Software , Ashland , OR ) . The highest RVP dilution yielding an infection rate of 7 to 15% was used for subsequent neutralization assays [32] . RVP neutralization assays were performed as previously described [32] . Briefly , RVPs were prepared according to the previously determined working dilution in a final volume of 25 µl of RPMI pH 8 . 0 complete medium . RVPs were then mixed with an equal volume of serum ( eight 3-fold serial dilutions in RPMI pH 8 . 0 complete medium starting at 1∶5 , tested in duplicate ) in 96-well plates and incubated on a shaker for 1 hour at room temperature . Infections were carried out as described above . The percentage of infected , GFP-positive cells for each serum concentration was plotted as percent infection versus log10 of the reciprocal serum dilution using Prism 5 . 0 ( GraphPad , La Jolla , CA ) . A sigmoidal dose response curve with a variable slope was then generated to determine the 50% neutralization titer , or NT50 – the serum dilution at which a 50% reduction in infection was observed compared to the positive ( no-serum ) control . Background GFP levels were subtracted from all measurements using a negative control ( no-RVP ) . Neutralization curves using reference sera ( polyvalent anti-DENV-1+2+3+4 serum code 02/186 , National Institute for Biological Standards and Control , UK ) were performed with serial 2-fold dilutions of all RVP lots to ensure that viral particles were neutralized according to the law of mass action [32] , [33] , such that serial dilutions of RVPs yielded the same NT50 , thus ensuring that the antibodies in the serum were in excess . Polyvalent serum was used in each neutralization assay to confirm neutralization against all 4 RVPs ( neutralization control ) . The RVP assay was standardized both at UC Berkeley and in Nicaragua . For each NT50 result , the absolute sum of squares ( ABSS ) and the coefficient of determination ( R2 ) of the non-linear regression were calculated . If the ABSS was >0 . 2 and/or the R2 was <0 . 9 , the data were excluded and repeated . An NT50 of <10 indicates a calculated NT50 value of <10 or the failure of the sera to neutralize at the lowest dilution by at least 50% . NT50 titers were independently calculated by two analysts . Thirty-nine participants who entered the cohort dengue-naïve and had experienced at least two DENV infections as determined by total antibody titer measurements ( ELISA ) were selected . As with antibody titration by ELISA , we used annual healthy serum samples and determined the NT50 for all four DENV serotypes . All participants in this subset had entered the cohort between 2004 and 2007 , and annual samples through 2011 were used , except for participants withdrawn from the study before then . The following rules for interpretation of the longitudinal NT50 data were established and implemented . For participants who had no evidence of a previous DENV infection ( i . e . , NT50 titers for all 4 DENV serotypes in all previous years were <10 ) , primary DENV infections were defined by seroconversion ( from NT50<10 to NT50≥10 ) to a specific serotype . For participants with evidence of prior DENV infection , secondary DENV infections were defined by seroconversion ( from NT50<10 to NT50≥40 ) or a ≥4-fold increase in NT50 ( fold-change was calculated as post-infection NT50/pre-infection NT50 ) . When several serotypes met the infection criteria during the same study year , the serotype with the highest NT50 fold-change was chosen . If the fold-change for more than one serotype was similar ( ±15% ) , an infection was assigned to the year but the serotype was recorded as unknown . If a symptomatic DENV infection with a given serotype was identified , no other infection with the same serotype was assigned throughout the years . If an inapparent DENV infection was identified , no other inapparent infection with the same serotype was assigned in later years . Interpretation of the DENV infection history of each participant over time was discussed by six authors to reach a consensus . For determination of the proportion of symptomatic DENV infections among total DENV infections , we only included symptomatic infections identified in participants who completed the study year and for whom paired annual samples were available ( 404 out of 448 symptomatic DENV infections ) . Statistical analyses were performed in STATA , version 12 ( StataCorp LP , College Station , TX ) . The binomial test was used to assess the distribution of DENV infections by sex . Chi-square and Fisher's exact tests were used to compare categorical variables among two ( or more ) independent groups . The Mann-Whitney U test was used to compare intervals between consecutive DENV infections .
A total of 5 , 541 children participated in the Pediatric Dengue Cohort Study from August 2004 to March 2011: 3 , 713 were enrolled at the onset of the study and 1 , 828 in subsequent years . We identified DENV infections during this period , corresponding to 7 dengue seasons . First , participants who met the WHO criteria for a suspected dengue case [2] and those with undifferentiated fever were evaluated for acute symptomatic DENV infection using molecular , virological , and serological diagnostic techniques ( see Methods ) . Second , inapparent DENV infections were identified using total DENV-specific antibody titers measured by Inhibition ELISA [29] , [30] in healthy annual blood samples from 8 annual collections ( 2004–2011 ) . The average number of annual samples contributed per participant was 5 . 3±2 . 1 ( Fig . S1A ) . DENV infections were stratified by study year; each year was delimited by two consecutive annual blood sample collections and encompassed a dengue season . Moreover , sequential first , second and third DENV infections were identified in participants who entered the study with no detectable anti-DENV antibodies ( “naïve” ) . As relatively few third infections were detected , an additional category was created to study post-second DENV infections by including 1 ) third infections in naïve participants , and 2 ) second and third infections experienced by children who entered the study with anti-DENV antibodies ( “non-naïve” ) . To identify first , second , third and post-second infections , participants who contributed two or more consecutive annual samples were included ( N = 5 , 082 ) . The average number of consecutive samples provided by these participants was 5 . 6±2 . 1 ( Fig . S1B ) . The average time interval between consecutive samples was 343±41 days ( Fig . S1C ) . Overall , we identified 448 symptomatic and 1 , 606 inapparent DENV infections ( Table 1 ) . Both symptomatic and inapparent infections were equally distributed by gender . However , repeat DENV infections tended to be more frequent in males ( chi-square test p = 0 . 060 ) ( Table 1 ) . We then analyzed the proportion of symptomatic DENV infections among all DENV infections . For this analysis , only participants with symptomatic DENV infections who had completed the study year were included ( n = 404 ) . The proportion of symptomatic DENV infections among all DENV infections was similar in females ( 20 . 8% ) and males ( 19 . 4% , chi-square test p = 0 . 447 ) . The mean age of infection was significantly higher ( p<0 . 001 ) , by 1 . 2 years , in symptomatic infections when compared to inapparent DENV infections ( Table 1 ) . We first examined the proportion of symptomatic infections among all DENV infections per study year . This proportion showed substantial differences , ranging from 4 . 9% in 2006–07 to 39 . 1% in 2009–10 ( “All infections” bars , Fig . 1A–G ) . Then , we analyzed the effect of infection number ( first , second , third and post-second ) on inapparent versus symptomatic DENV infection outcome . For each study year , trend analyses were performed with first , second and post-second DENV infections , as the number of third infections was limited . For all study years but one , the proportion of symptomatic DENV infections was similar in first , second , and post-second infections ( Fisher's exact test , p>0 . 05 , Fig . 1A–G ) . In 2008–09 , no symptomatic second infections and very few symptomatic post-second infections were identified when compared to symptomatic first infections ( Fisher's exact test , p = 0 . 003 ) ( Fig . 1E ) . Overall , this analysis suggests that , in this study , inapparent versus symptomatic outcome is similar in first , second and post-second DENV infections . For participants with repeat DENV infections , we then examined whether symptomatic versus inapparent outcome of a prior infection influences outcome of a subsequent infection . To this end , the proportion of symptomatic infections was calculated given the outcome of the previous infection . No significant difference was observed , as the proportion of symptomatic DENV infection was 24 . 9% when the previous infection was inapparent ( N = 293 ) and 23 . 5% when the previous infection was symptomatic ( N = 34 ) ( chi-square test p = 0 . 859 ) . We then evaluated the effect of the time interval between infections on repeat DENV infections . The interval between two consecutive infections was defined as the number of seasons between the infections . For instance , the interval between an infection in 2005–06 and another infection in 2008–09 is 3 years . In total , 341 intervals between DENV infections were calculated . The mean interval was 2 . 4 years . Next , we stratified the intervals between infections with respect to the outcome of both the prior and the subsequent infection . Four different infection sequences were thus defined: an inapparent DENV infection followed by another inapparent infection ( inapparent-to-inapparent ) or by a symptomatic infection ( inapparent-to-symptomatic ) , and a symptomatic DENV infection followed by an inapparent infection ( symptomatic-to-inapparent ) or another symptomatic infection ( symptomatic-to-symptomatic ) . The mean interval was calculated for each of the four groups ( Fig . 2A ) . Notably , the inapparent-to-inapparent infection mean interval was significantly shorter than the inapparent-to-symptomatic infection interval ( 2 . 2 versus 2 . 7 years , Mann-Whitney U test p = 0 . 021 ) ; all other pairwise comparisons were not significant . We further stratified the infection sequences by infection number . Specifically , for participants who entered the cohort dengue-naïve , infection sequences were divided into “first-to-second” and “second-to-third” DENV infections . In the “first-to-second” group , the inapparent-to-inapparent infection interval was again significantly shorter than the inapparent-to-symptomatic infection interval ( 1 . 8 versus 2 . 6 years , Mann-Whitney U test p = 0 . 018 ) ( Fig . 2B ) . The other pairwise comparisons were not significant . The symptomatic-to-symptomatic infection sequences were not included in the analysis as no “second-to-third” such sequence was observed . Interestingly , no difference was observed when comparing inapparent-to-inapparent and inapparent-to-symptomatic infection intervals for “second-to-third” infection sequences ( 2 . 7 versus 2 . 5 years , p = 0 . 692 ) . Moreover , the inapparent-to-inapparent infection interval was significantly shorter in “first-to-second” ( 1 . 8 years ) than in “second-to-third” infection sequences ( 2 . 7 years , Mann-Whitney U test p = 0 . 005 ) . However , this observation was limited by the small number of “second-to-third” infections sequences analyzed ( 11 inapparent-to-inapparent and 13 inapparent-to-symptomatic ) . To extend this observation , we created a new group of infection sequences by adding to the “second-to-third” sequences those infections observed in participants who entered the cohort non-dengue-naïve . This new group was termed “other infection sequences” as it includes all possible DENV infection sequences except the “first-to-second” infection group . Notably , no difference was observed between the inapparent-to-inapparent and inapparent-to-symptomatic infection intervals within this group ( Fig . 2C ) . Furthermore , when comparing the inapparent-to-inapparent infection interval between the “first-to-second” and the “other infection sequences” groups , the former was found to be significantly shorter ( 1 . 8 versus 2 . 7 years , Mann-Whitney U p<0 . 001 ) ( Fig . 2B–C ) . The symptomatic-to-symptomatic infection sequences were not included in this analysis due to the small number of observations ( “first-to-second” N = 5; “other infection sequences” N = 5 ) . Taken together , these show that the interval between two inapparent infections is significantly shorter than the inapparent-to-symptomatic infection interval , but only when considering the first and second DENV infections of a given participant . We then undertook a longitudinal analysis of DENV serotype-specific neutralizing antibody titers in a subset of cohort participants . The objective of this analysis was to examine the feasibility of reconstructing participants' DENV immune history using a Reporter Viral Particle ( RVP ) flow cytometry-based DENV neutralization assay [32] and to substantiate the results obtained with Inhibition ELISA by measuring neutralizing antibodies instead of total anti-DENV antibodies . This assay yields reproducible serotype-specific neutralization titers that are in agreement with Plaque Reduction Neutralization Test ( PRNT ) results [32] . First , we examined the ability of the 50% neutralization titer ( NT50 ) changes between pre- and post-infection annual samples to detect symptomatic DENV infections and to identify the correct DENV serotype in a subset of 27 confirmed symptomatic infections with serotype information available from RT-PCR and/or virus isolation . The pre- to post-infection fold-change in NT50 was calculated for each DENV serotype . Using the highest NT50 fold-change as an indicator , 26 out of 27 DENV serotypes were correctly identified ( Fig . S2 ) . In one additional case ( participant M , Fig . S2 ) , taking into account the participant's immune history allowed for the identification of the infecting serotype ( DENV-3 ) . In this case , the participant had experienced an inapparent infection with DENV-2 prior to the symptomatic episode . The NT50 fold-change was highest for DENV-2 but , consistent with the interpretation rules we had established , the infecting serotype was recorded as DENV-3 , which had the second highest NT50 increase . Second , we analyzed longitudinal data from 39 cohort participants to determine their DENV-specific immunological history by compiling symptomatic and inapparent DENV infections as detected in consecutive annual samples ( see Methods for specific rules ) . Longitudinal NT50 titers for two participants are shown in Figure 3 . Both participants displayed an NT50<10 against all 4 serotypes in their initial sample and were therefore considered dengue-naïve . Participant A apparently experienced an inapparent DENV-2 infection in 2005–06 followed by an inapparent DENV-4 infection in 2006–07 . Subsequently , NT50 titers did not display any major changes until 2010 , when titers for all four serotypes increased more than 4-fold . However , the most likely infecting serotype was determined to be DENV-3 as the increase in NT50 against DENV-3 was the greatest , aside from DENV-2 , which had caused the first infection . In fact , this participant experienced a symptomatic DENV-3 infection in 2009–10 as determined by RT-PCR and viral isolation using acute and convalescent samples from the period of illness . Participant B experienced 3 inapparent DENV infections: DENV-1 in 2005–06 , DENV-2 in 2007–08 and DENV-3 in 2009–10 . Overall , 75 inapparent DENV infections were detected among the 39 participants analyzed ( Table S2 ) . For most infections ( 73/75 ) , the likely infecting serotype was identified . For the remaining two , a comparable fold-change in NT50 titers was observed for two serotypes , making it difficult to assign an infecting serotype . Finally , we compared DENV serotype circulation in each study year as determined by neutralization assay using annual samples to symptomatic DENV infections detected in the entire cohort by RT-PCR and/or virus isolation . Serotype circulation was similar using both approaches , showing that the circulating serotype ( s ) cause both inapparent and symptomatic DENV infections and further validating the neutralization method ( Fig . S3 ) . The only striking difference was DENV-4 circulation in 2006–07 , 2007–08 and 2009–10 , which only caused inapparent infections . These data are consistent with limited PRNT data that we obtained as part of a study of DENV neutralizing antibodies in a random 10% of the cohort from 2004 to 2007 and in a subset of inapparent infections in different individuals each year from 2004 to 2008 , where inapparent DENV-4 infections were also identified in 2006–07 and 2007–08 ( M . J . Vargas , A . Balmaseda , E . Harris , unpublished results ) . Using the same approach as for total antibody titers above , the intervals between consecutive DENV infections were determined in the subset of cohort participants examined using the neutralization assay . The mean interval between two DENV infections was 2 . 41 years ( N = 54 ) . Despite the fact that the neutralization titer dataset contained approximately 6 times fewer consecutive infection sequences than the ELISA dataset from the entire cohort , the value obtained in the neutralization subset was similar to the mean interval determined using total antibody titer ( 2 . 35 years ) . We then stratified the infection sequences by infection outcome and infection number . Only inapparent-to-inapparent and inapparent-to-symptomatic infection sequences were compared , as the number of symptomatic-to-inapparent infections was small ( N = 4 ) and no symptomatic-to-symptomatic infection sequences were observed . When comparing all intervals , the inapparent-to-inapparent infection interval was significantly shorter than the inapparent-to-symptomatic infection interval ( Mann-Whitney U test p = 0 . 025 ) ( Fig . 4A ) . However , when we stratified by infection number , this difference was only observed in “first-to-second” subset ( Mann-Whitney U test p = 0 . 003 , Fig . 4B ) and not when considering “second-to-third” infection sequences ( Fig . 4C ) . These results corroborate the findings obtained with consecutive DENV infection interval using total antibody titers in the entire cohort .
In this study , we analyzed several determinants of inapparent versus symptomatic DENV infection , taking advantage of our long-term Pediatric Dengue Cohort Study in Managua , Nicaragua . Data from 1 , 606 inapparent and 448 symptomatic DENV infections were collected over 7 years using annual total anti-DENV antibody titers as measured by Inhibition ELISA and “enhanced” passive surveillance of febrile cases , respectively . Overall , symptomatic DENV infections were equally distributed by gender but more frequent in older children . The proportion of symptomatic DENV infections among all DENV infections varied substantially across study years but was not significantly affected by infection number ( i . e . , first , second , third , or post-second infections ) . In participants with documented repeat DENV infections , the outcome of a previous DENV infection did not influence the outcome of the subsequent infection; however , the time interval between two consecutive infections did . In fact , the interval between two inapparent DENV infections was significantly shorter that the interval between an inapparent and a symptomatic infection . However , this result was only observed when considering the first and second DENV infections of a given participant . Moreover , this finding was confirmed using a flow cytometry-based neutralization assay to quantify serotype-specific anti-DENV neutralizing antibodies in a subset of cohort participants . The proportion of symptomatic DENV infections among total infections was found to be similar in females and males , consistent with observations in other studies [3] , [14] . However , age played a role in influencing symptomatic outcome , as symptomatic DENV infections tended to occur more frequently in older children . Interestingly , this effect was not observed in the Kamphaeng Phet ( Thailand ) cohort [3] . The most striking determinant of infection outcome was the study year . We had previously reported large variations in the proportion of symptomatic DENV infections in the first four dengue seasons of the Pediatric Dengue Cohort Study ( 2004–05 to 2007–08 ) [15] . Here , we extended this analysis through 2010–11 and found even more dramatic variations , from ∼5–6% in 2004–05 and 2006–07 to almost 40% in 2009–10 and 2010–11 . Similar temporal variations have been reported in other studies in Peru [34] and Thailand [3] , [4] , [35] . The factor ( s ) driving these differences in our Nicaraguan cohort are not completely known , although in 2007–08 a clade replacement within DENV-2 is thought to have contributed to the higher proportion of symptomatic infections [24] , and in 2009–10 the concurrent H1N1 influenza pandemic may have played a role [23] . Overall , we did not observe a correlation between circulating serotypes and infection outcome , except for DENV-4 , which caused mostly inapparent infections . In the cohort study from 2004 to 2011 , only one DENV-4 symptomatic infection was reported . However , in the subset of 39 participants who were analyzed using the serotype-specific neutralization assay , 9 inapparent DENV-4 infections were detected , suggesting that DENV-4 does circulate in Managua but rarely causes symptomatic infections . Conventionally , DENV infections have been defined as primary or secondary depending on the immune response profile in acute and convalescent samples [2] . No distinction is usually made between second , third and fourth DENV infections , as differences in the immune response between these categories are notoriously difficult to determine . Studying specifically first versus second versus third versus fourth DENV infections requires long-term cohort studies that capture both inapparent and symptomatic infections in the same individuals over time . Here , we report inapparent versus symptomatic outcome in first , second and third DENV infections . As the number of third infections was relatively small , we also analyzed outcome in post-second infections . Interestingly , when stratified by study year , the proportion of symptomatic DENV infections was similar in first , second , third , and post-second infections . The data provided here about post-second and third infections are important , as models suggest that post-second infections could impact dengue dynamics , overall force of infection , and incidence rates of severe dengue disease [36] . However , to date , few models have been able to incorporate information about infection number for lack of specific data about second versus third versus fourth DENV infections . In addition , there are implications for vaccine development . If , in fact , there is substantial symptomatic disease in post-second infections , then tetravalent or at least trivalent seroconversion after vaccination would be crucial for effective vaccine protection . Both seminal observations by Sabin [5] and epidemiological reports [12] , [13] , [37] suggest that the time interval between consecutive DENV infections plays a role in infection outcome and severity . Here , we analyzed the time interval between repeat DENV infections and evaluated its impact on inapparent versus symptomatic outcome . As healthy blood samples were collected annually in this study , the intervals between consecutive DENV infections were calculated as integers representing annual increments . The mean interval between two DENV infections in our entire dataset was 2 . 4 years . We found that after an inapparent DENV infection , the interval to a subsequent inapparent DENV infection was significantly shorter than the interval to a subsequent symptomatic DENV infection ( 2 . 2 versus 2 . 7 years , p = 0 . 021 ) . Similar numbers were obtained when the preceding infection was symptomatic , although the number of observations was small and the difference was not significant . Interestingly , the shorter inapparent-to-symptomatic infection interval was only observed when , for a given participant , the preceding infection was his/her first DENV infection and the subsequent infection the second . In this case , the inapparent-to-inapparent interval was 1 . 8 years versus 2 . 6 years for inapparent-to-symptomatic infection . These results suggest that the immunity induced by a first infection with DENV protects against a second symptomatic infection for approximately 2 years . Then , immunity wanes and is no longer protective . However , we cannot exclude that confounding factors such as age and yearly serotype-specific infection rates may contribute to the observed differences between inapparent-to-inapparent and inapparent-to-symptomatic intervals . These results are consistent with the time interval of cross-protection estimated between DENV-1 and DENV-2 infections in Nicaragua in 2005–08 [23] . These findings are also consistent with Sabin's observations , although the protection window of a few months described in his experimental study is shorter [5] . To the best of our knowledge , this is the first published report measuring the specific time interval of cross-protection prior to a subsequent DENV infection in the context of natural DENV infections . It is well-established that secondary heterotypic DENV infection is the most important risk factor for severe disease [7]–[11] . In our cohort study , a similar effect is observed: 3 . 0% of secondary DENV infections were classified as DHF/DSS as compared to only 0 . 8% of primary infections . However , the total number of DHF/DSS cases identified in the study ( n = 42 ) was too small to stratify them by first versus second versus third ( or post-second ) infections and to evaluate the impact of the time interval between consecutive DENV infections on disease severity . The dengue plaque reduction neutralization test ( PRNT ) is currently considered the “gold standard” to quantify serotype-specific anti-DENV neutralizing antibodies , although it has not been well-standardized across difference laboratories in terms of reagents and testing conditions [38]–[40] . However , the size and longevity of the Pediatric Dengue Cohort Study make it logistically unfeasible to use PRNT for annual serological testing . Here , we used two serological techniques . First , to measure total anti-DENV antibodies in the large number of annual samples collected , we used the Inhibition ELISA [29] , [30] . The Inhibition ELISA has been previously evaluated in Nicaragua and showed a sensitivity of 98 . 9% and a specificity of 100% as compared to the Hemagglutinin Inhibition assay [29] . Although the Inhibition ELISA is a fast and reliable technique , it does not provide serotype information nor does it specifically measure neutralizing anti-DENV antibodies . Thus , we used a second serological assay: the Reporter Viral Particle ( RVP ) flow cytometry-based DENV neutralization assay in a subset of participants . This technique has been previously evaluated and generate neutralization titers that are in good statistical agreement with PRNT [32] . A thorough quality control procedure was implemented at all steps of the assay from reagent control to data analysis . Specific rules were established to infer DENV infections from the annual sample neutralization titers . Using this set of rules , all symptomatic DENV infections identified in the subset of cohort participants were correctly captured using the RVP-generated neutralization titers . Furthermore , comparison of the serotype identified by RT-PCR and/or virus isolation and the serotype identified using NT50 values was 100% concordant . However , the throughput of the flow cytometry-based neutralization technique is limited compared to Inhibition ELISA , and we were only able to use it to analyze a subset of samples . The neutralization antibody data was used to confirm our findings on the time interval between repeat DENV infections . Notably , the intervals calculated using the neutralization assay closely matched those obtained using Inhibition ELISA data . One of the limitations of this study is that serotype information is available for only a subset of the inapparent DENV infections – those processed using the RVP-based neutralization assay . We are currently expanding the number of annual samples processed using this technique . This will allow us to address several unanswered questions regarding inapparent versus symptomatic DENV infection outcome , including the impact of DENV serotype and the sequence of DENV serotypes on outcome and the effect of the magnitude and breadth of pre-infection neutralizing titers on infection outcome . Another limitation is the particular epidemiological context of dengue epidemics in Nicaragua . In contrast to hyperendemic areas in Asia where all four DENV serotypes circulate simultaneously , in Nicaragua one serotype predominates in each dengue season [22]–[24] . Moreover , a substantial amount of symptomatic infections reported in this study occurred in 2009–10 and 2010–11 , when DENV-3 was the main circulating serotype , and this could conceivably influence the determinants of symptomatic versus inapparent DENV infection outcome . Future studies will show if these determinants , in particular the time interval between consecutive DENV infections , are comparable in a hyperendemic context . Collectively , our results shed light on the factors influencing inapparent versus symptomatic DENV infection outcome . We show that while sex and infection number did not impact infection outcome , age and study year did . In the context of our long-term Pediatric Dengue Cohort Study , we were able to investigate participants with repeat DENV infections . Our results suggest that infection number ( i . e . , first , second , third , or post-second DENV infection ) does not significantly impact inapparent versus symptomatic outcome , at least in our study . However , the time interval between a first and a second DENV infection plays a significant role in infection outcome , as a shorter interval between infections is associated with inapparent outcome . These results highlight the role of heterologous cross-protection between natural DENV infections and the importance of prospective cohort studies to study repeat DENV infections . | The four serotypes of the mosquito-borne dengue virus ( DENV ) infect an estimated 100 million humans annually , resulting in tens of millions of dengue cases and hundreds of thousands of cases of severe disease . However , infection with DENV does not always lead to clinical signs , and a large proportion of DENV infections are inapparent . Here , we studied the factors that influence whether a DENV infection is inapparent or symptomatic . Data from over 2 , 000 DENV infections ( ∼1 , 600 inapparent and ∼400 symptomatic ) were collected during 7 years from an ongoing prospective cohort study of children in Managua , Nicaragua . We show that whether a person is infected for the first , the second , or the third time with different DENV serotypes , the proportion of symptomatic infections is similar . However , the proportion of symptomatic infection varied substantially across study years , and symptomatic infections tended to happen in older children when compared to inapparent infections . We also show that if a second DENV infection happens within a period of ∼2 years after the first infection , the second infection is more likely to be inapparent . However , if the time interval between first and second DENV infections is longer , this protection wanes and the infection is likely to be symptomatic . These findings are important for the modeling of dengue epidemics and the development of new vaccines . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"dengue",
"fever",
"neglected",
"tropical",
"diseases"
] | 2013 | Symptomatic Versus Inapparent Outcome in Repeat Dengue Virus Infections Is Influenced by the Time Interval between Infections and Study Year |
Intercellular communication is critical for the survival of unicellular organisms as well as for the development and function of multicellular tissues . Cell-to-cell signaling is also required to develop the interconnected mycelial network characteristic of filamentous fungi and is a prerequisite for symbiotic and pathogenic host colonization achieved by molds . Somatic cell–cell communication and subsequent cell fusion is governed by the MAK-2 mitogen activated protein kinase ( MAPK ) cascade in the filamentous ascomycete model Neurospora crassa , yet the composition and mode of regulation of the MAK-2 pathway are currently unclear . In order to identify additional components involved in MAK-2 signaling we performed affinity purification experiments coupled to mass spectrometry with strains expressing functional GFP-fusion proteins of the MAPK cascade . This approach identified STE-50 as a regulatory subunit of the Ste11p homolog NRC-1 and HAM-5 as cell-communication-specific scaffold protein of the MAPK cascade . Moreover , we defined a network of proteins consisting of two Ste20-related kinases , the small GTPase RAS-2 and the adenylate cyclase capping protein CAP-1 that function upstream of the MAK-2 pathway and whose signals converge on the NRC-1/STE-50 MAP3K complex and the HAM-5 scaffold . Finally , our data suggest an involvement of the striatin interacting phosphatase and kinase ( STRIPAK ) complex , the casein kinase 2 heterodimer , the phospholipid flippase modulators YPK-1 and NRC-2 and motor protein-dependent vesicle trafficking in the regulation of MAK-2 pathway activity and function . Taken together , these data will have significant implications for our mechanistic understanding of MAPK signaling and for homotypic cell–cell communication in fungi and higher eukaryotes .
Intercellular communication is critical for the survival of simple unicellular organisms such as bacteria and yeasts and is central for the development and function of multicellular plant and animal systems [1]–[4] . Cell-cell signaling and somatic cell fusion is also required to develop the interconnected mycelial network characteristic of filamentous fungi [5] . This feature is important for the fitness of the fungal colony by the shared use of information , nutrients and organelles between individual cells [6] , [7] . Consequently , hyphal anastomosis is critical for host colonization and symbiotic interactions as well as for virulence of pathogenic species [8]–[12] . Hyphal fusion is comparable to homotypic cell fusion between genetically identical cells of higher eukaryotes , which results in the formation of multinucleate syncytia [13] , [14] . Important examples for human biology are myoblast fusion during muscle differentiation , trophoblast fusion during placental development and osteoclast fusion during bone formation . Thus , fungal self-signaling may provide a powerful model for understanding molecular mechanisms of homotypic cell communication during animal and human tissue development . In the ascomycete model mold Neurospora crassa , an unknown chemical ligand mediates chemotropic communication between genetically identical cells . Germinating spores mutually attract each other and subsequently fuse to generate an interconnected network of multinucleate cells that form the mycelial colony [5] , [15] . This process of self-signaling is based on the oscillatory recruitment of the NRC-1–MEK-2–MAK-2 mitogen activated protein kinase ( MAPK ) cascade ( homologous to the Ste11p-Ste7p-Fus3p mating pathway in budding yeast ) and of SOFT , a protein of unknown molecular function , to the opposing tips of communicating germlings [16] , [17] . The rapid alternation of these two different physiological states of “homing” cells likely reflects signal response and delivery , respectively . Although these findings resulted in a first qualitative model to describe the excitable behavior of the MAK-2 module [18] , our understanding of oscillatory MAK-2 signaling is hampered by the fact that most components of the signaling machinery – including the postulated secreted signal and its cognate receptor ( s ) , regulators of the MAPK cascade as well as most MAK-2 targets – are unknown . N . crassa and other filamentous fungi possess G-protein coupled receptors , heterotrimeric G-proteins , STE20-related kinases , components of the cAMP machinery and Ras/Rho-type GTPase modules known to function upstream or in parallel of MAPK signaling in fungi and higher eukaryotes [19]–[22] . However , mutant analyses indicate that individual deletions of these components are dispensable for vegetative cell communication ( summarized in [23] ) . We hypothesized that redundant functions between the mentioned proteins require additional approaches to classical mutant hunts in order to dissect MAK-2 signaling . In this study , we used a proteomics approach that allowed the identification of STE-50 as regulatory subunit of the Ste11p homolog NRC-1 and HAM-5 as cell-communication-specific scaffold protein of the MAPK cascade . Moreover we defined a network of proteins , consisting of two Ste20-related kinases , the small G-protein RAS-2 and the adenylate cyclase capping protein CAP-1 , whose signals converge on the MAK-2 pathway .
A high-quality interactome of the MAK-2 MAPK module was generated by affinity purification experiments coupled to mass spectrometry ( AP-MS ) with strains expressing functional GFP-fusion proteins of the three kinases of the MAK-2 cascade ( Table S1 ) . Each of the three bait proteins recovered the tripartite kinase cascade with high stringency in the two biological replicate purifications performed , indicating the suitability of the approach . Moreover , we identified two proteins , STE-50 and HAM-5 , which displayed tight interactions with all three kinases ( Figure 1 A ) . The remaining hits that associated with all three kinases were only identified with poor coverage and/or in one of the two purifications and thus could constitute contaminants . However , among them were PPG-1 , PP2A-A , HAM-3 and MOB-3 , components of the striatin-interacting phosphatase 2A and kinase ( STRIPAK ) complex that is required for fungal cell-cell signaling [24]–[29] . This suggests that at least some of the additional hits may represent dynamically , and thus weakly interacting components . Thus , we assayed available mutants defective for additional candidate proteins that associated with the three kinases for tropic interaction defects and determined that the casein kinase 2 heterodimer CKA/CKB-1 , the serine/threonine kinase YPK-1 , type V myosin/NCU01440 , and the hypothetical protein NCU06265 ( designated HAM-13 ) were required for proper cell communication . Consistent with these defects , we also detected reduced MAK-2 phosphorylation levels in cell extracts of these five mutant strains ( Table 1; Figure S1 ) . The combined proteomics data and mutant characteristics indicate a specific involvement of these proteins in cell-cell communication despite the fact that they were identified only with poor coverage in the AP-MS analysis . However , further experiments are required to determine their mechanistic mode of action and to exclude indirect effects of the mutations on MAK-2 pathway functionality . In addition , mutants defective for the hypothetical proteins NCU00627 , NCU02606 , NCU02972 , and NCU08957 that also interacted with the entire kinase cascade displayed inconspicuous communication patterns and were thus not tested for MAK-2 phosphorylation , because they likely represented contaminants . Interestingly , we also determined that MAK-2 , yet not the two upstream kinases interacted with multiple components of the nuclear import/export machinery ( i . e . importin alpha , importin beta-1 , importin beta-3 , exportin-1 , nuclear pore protein NCU01702; Table S1 ) and the two transcriptional regulators PP-1 and RCO-1 , which were previously identified as key effectors of the MAK-2 pathway [30]–[32] . Budding yeast Ste50p functions as adaptor protein of the MAP3K Ste11p that connects heterotrimeric G-proteins and small GTPases with various MAPK cascades [33] , [34] . This is achieved through the modular structure of Ste50p . The protein consists of an N-terminal protein interaction domain called the sterile alpha motif ( SAM ) [35] , [36] and a C-terminal Ras association ( RA ) domain that can bind to small Ras and Rho-type GTPases and is required for membrane delivery of the Ste11p/Ste50p complex [37] , [38] . Homologs in filamentous fungi are not as well characterized , but have been proposed to function as scaffold proteins in MAPK cascades [39]–[42] . Yeast two hybrid ( Y2H ) assays confirmed the physical interaction of STE-50 with itself and with NRC-1 , yet not MEK-2 and MAK-2 , through the SAM domains present in the two proteins ( Figure 1 B , C; Figure S2 ) . The presence of a stable STE-50/NRC-1 complex was further supported by the reciprocal AP-MS analysis of GFP-STE-50 interacting proteins , which revealed a stable interaction of STE-50 with NRC-1 , but not other components of the MAPK cascade ( Figure 1 A ) . Strains expressing gfp-ste-50 under its native promoter complemented the defects of the deletion strain , but barely showed any fluorescence , which was consistent with the weak expression profile of its binding partner NRC-1 [17] . Thus , we analyzed strains expressing the fusion construct under the control of the Pccg-1 promoter . GFP-STE-50 displayed a subcellular dynamics similar to that previously described for the three kinases of the MAK-2 cascade [16] , [17] and accumulated at the future fusion site after the two germlings had established physical contact ( Figure 2 A ) . Consistent with the expression profile and localization pattern described for NRC-1 [17] , we observed only weak cytoplasmic localization of GFP-STE-50 in mature hyphae and exclusion of the fusion protein from nuclei . However , GFP-STE-50 accumulated at septa and contact sites of fusing hyphae within an established colony ( Figure 2 B ) . We determined that a Δste-50 deletion mutant fully phenocopied all defects described for defective MAK-2 signaling [31] , [43] , [44] and lacked detectable MAK-2 kinase activity ( Table 1; Figure 2 C; Figure S3 ) . Based on these characteristics of the mutant and the tight interaction of STE-50 with NRC-1 , we hypothesized that STE-50 may function as regulatory subunit of the MAP3K . To test this , we expressed a recently generated [17] , constitutive active NRC-1 version NRC-1 ( P488S ) in Δste-50 . This construct fully complemented all defects of the deletion strain and confirmed our hypothesis ( Figure 2 B ) . Ham-5 was originally identified in a genetic screen for N . crassa mutants that fail to undergo cell fusion [30] . Its closest homolog , Podospora anserina IDC1 , is required for NADPH-oxidase-dependent nuclear accumulation of the cell wall integrity MAPK Mpk1 [45] , but the mechanistic basis underlying this observation remains unresolved . MAK-2 activity was reduced to ca . 1/10 of wild type in the Δham-5 mutant ( Table 1; Figure S3 ) , indicating significantly compromised , but not abolished MAK-2 pathway function in this strain . We observed increased mycelial extension rates of Δham-5 when compared to Δste-50 or Δmak-2 , residual tropic interactions ( ≤10% of wild type ) and the retained capacity to produce low amounts of infertile protoperithecia ( ca . 20% of wild type; Figure S3 ) . Co-immunoprecipitation experiments were used to confirm the interaction of HAM-5 with all three kinases of the MAK-2 pathway Figure S2 ) . In order to further dissect the physical interaction pattern of these proteins , we performed Y2H tests . In these assays , we determined that HAM-5 , which contained seven N-terminally located WD40 repeats and two short coiled coil regions in the C-terminal region , was able to homo-dimerize and also interacted with STE-50 and with MAK-2 , yet not with MEK-2 and NRC-1 ( Figure 1 B , C; Figure S2 ) . Functional Pccg-1-driven HAM-5-GFP displayed the predicted dynamic subcellular localization during germling communication and strictly co-localized with MAK-2 as dynamically forming intracellular complexes that associated with the communicating tips of the two cells with an oscillation period of three to five minutes ( Figure 3 A , B ) . At least some of the tip-associated signal was generated by recruitment of cytosolic HAM-5- and MAK-2-containing puncta to the apex ( Figure 3 C ) . Diffuse HAM-5-GFP label was excluded from nuclei , and bright HAM-5 puncta were not obviously associated with nuclei . However , we detected weakly labeled HAM-5 puncta that decorated nuclear envelopes ( preferentially marking those nuclei that localized closely to the cell tip ) simultaneously with the first appearance of tip-associated HAM-5 signal . ( Figure 3 D ) . Dynamic co-recruitment of HAM-5 and MAK-2 to communicating hyphal tips , septa and intracellular puncta was also observed in the mature colony ( Figure 4 A , Figure S4 ) . Moreover , HAM-5 and SOFT oscillated with opposing recruitment phases in communicating hyphae ( Figure S4 ) , indicating that the molecular machinery required for germling communication is also operating during hyphal anastomosis within the mature colony . In contrast to MAK-2 [17] , HAM-5 was not detected at the apex of non-communicating hyphae ( Figure 4 B ) . We therefore asked if MAK-2 activity is required for HAM-5 aggregation during cell signaling . When we localized HAM-5-GFP in Δmak-2 germlings , we observed the presence of HAM-5-GFP puncta , which remained stable over long time periods ( Figure 4 C ) . Moreover , we noticed that intracolonial Δmak-2 hyphae that had ( by chance ) established physical contact grew in parallel over longer distances with HAM-5-GFP puncta formed primarily in one of the two hyphae , while septal pores were strongly labeled in the second cell ( Figure 4 C ) . Quantification of this phenomenon revealed that 72±7% ( n = 141 ) of the analyzed intracolonial hyphal pairs displayed this HAM-5 distribution . Thus , MAK-2 is essential to regulate the subcellular dynamics of HAM-5 , but complex formation does not require the MAPK . We reasoned that the interaction pattern of HAM-5 with multiple MAK-2 pathway components and its localization dynamics were consistent with a scaffold function of HAM-5 for the MAK-2 cascade . We tested this hypothesis by performing co-immunoprecipitation experiments ( Figure 4 D ) . NRC-1 and MAK-2 co-precipitated when tagged versions of both proteins were co-expressed in wild type , while we were unable to detect interactions between the two kinases in a Δham-5 background , indicating that HAM-5 is critical for maintaining the integrity of the kinase cascade . Our AP-MS analysis identified several proteins that specifically interacted with NRC-1 , yet not MEK-2 or MAK-2 and thus could constitute upstream components of the MAK-2 pathway . Among them were the MAP4K STE-20 , the small GTPase RAS-2/SMCO-7 and the capping protein CAP-1/NCU08008 of the adenylate cyclase ( AC ) complex ( Table S1 ) . A fully functional STE-20-GFP fusion construct ( Figure S5 ) recovered STE-50 , NRC-1 , HAM-5 and RAS-2 in reciprocal AP-MS experiments ( Figure 1 A ) , and Y2H assays confirmed physical interactions of STE-20 with HAM-5 , NRC-1 and STE-50 ( Figure 1 B , C; Figure S2 ) . Moreover , NRC-1 and weakly also STE-50 interacted with RAS-2 through the Ras-association domains present in both proteins in Y2H tests . We did not detect any Y2H interaction of STE-50 and NRC-1 with the second Ras-type GTPase , RAS-1/NCU08823 ( also called BAND; [46] ) , present in N . crassa , underscoring the specificity of the MS and Y2H analyses . We also detected MST-1 , a recently identified accessory Ste20-related kinase of the septation initiation network , which localized to spindle pole bodies and septa [47] , [48] as NRC-1- and MEK-2-interacting protein by AP-MS and Y2H analysis ( Table S1; Figure 1 B ) . However , communication frequencies and MAK-2 activities of Δmst-1 and of a generated Δste-20;Δmst-1 double mutant indicated that MST-1 is not critically required for MAK-2 signaling ( Table 1 ) . Consistent with an involvement in cell-cell communication , we found that STE-20-GFP formed membrane-associated apical crescents in mature hyphae , was enriched in a stable manner at both communicating germling tips and localized at the contact point of interacting cells in addition to its association with septa in germlings as well as mature hyphae ( Figure 5 A; Figure S5 ) . Moreover , a functional GFP-RAS-2 construct distributed along the entire plasma membrane in germlings and hyphae and localized to the contact point of interacting germlings ( Figure 5 B; Figure S5 ) . As predicted for components that signal upstream of the MAK-2 cascade , we observed reduced MAK-2 activities and cell communication frequencies in Δste-20 , in Δras-2 , the previously described ras-2 allele smco-7 [49] , and in two mutants affecting the predicted RAS-activating GDP-GTP-exchange factor CDC-25 ( Table 1; Figure 5 C ) . Only residual levels of germling communication and MAK-2 activity were detected in a Δste-20;Δras-2 double mutant , highlighting the joint importance of RAS-2 and STE-20 for self-signaling . The MAK-2 pathway is also induced by other external stimuli , such as reactive oxygen species [17] , [28] , and we asked if the identified proteins were also required for stress-induced activation of MAK-2 . We determined that H2O2-induced activation of the MAK-2 pathway was abolished in Δras-2 and in the Δste-20;Δras-2 double mutant ( Figure 5 D ) , indicating that both proteins are general components of the MAK-2 pathway . We also tested for binding of these components to the AC capping protein CAP-1 and detected a positive Y2H interaction with STE-50 , but not with any of the other proteins of the MAK-2 pathway ( Figure 1 B ) . This may indicate that our AP-MS-based identification of CAP-1 in precipitates of STE-50 , NRC-1 and STE-20 ( Table S1; Figure 1 A ) was mediated through STE-50 . Nevertheless , we observed altered cell communication frequency and reduced MAK-2 activity in Δcap-1 , consistent with a functional involvement of CAP-1 in MAK-2 signaling ( Table 1 ) . Significantly , we also detected reduced tropic germling interactions in the AC mutant Δcr-1 ( Figure 5 C ) , indicating that cAMP signaling is involved in , but is not essential for cell-cell communication . These data are consistent with a previous report , which showed that cr-1 germlings are able to generate conidial anastomosis tubes , although no quantitative analysis was performed in this study [50] .
Despite considerable progress in recent years , our mechanistic understanding of oscillatory MAK-2 behavior during homotypic cell communication is hampered by the fact that many components of the signal transduction machinery are still unknown and that the molecular functions of proteins known to be required for signaling are only poorly understood . One important finding of this study is the identification of STE-50 and HAM-5 as central components of the MAK-2 pathway ( Figure 6 ) . We propose STE-50 as tightly associated , regulatory subunit of NRC-1 and HAM-5 as scaffold protein of the MAK-2 cascade . Based on our data , STE-50 may have both NRC-1-activating as well as targeting functions , and we currently cannot rule out any of these hypotheses . Nevertheless , RAS-2 interacts with both NRC-1 and STE-50 in Y2H assays , and thus STE-50 may not be essential for membrane targeting , yet full complementation of Δste-50 by expression of constitutive active NRC-1 ( P488S ) indicates that STE-50 is critical for activation of the MAP3K . We do not have any evidence for a scaffold function of STE-50 in N . crassa , contrasting data obtained in other filamentous fungi , which have suggested interactions of STE-50 homologs with other kinases in addition to the MAP3K [40] , [42] , [51] . Central for our understanding of scaffold proteins is the archetypical MAP kinase scaffold Ste5p of the yeast mating pathway [52] , [53] . However , Ste5p is restricted to budding yeast and close relatives [20] , [54] . In contrast , homologs of HAM-5 are detected in all sequenced members of the Pezizomycotina subphylum ( the group of filament-forming ascomycete fungi ) , but are absent from the genomes of unicellular ascomycete fungi ( Figure S6 ) . Thus , we propose HAM-5 as scaffold of the N . crassa MAK-2 cascade and of homologous MAPK modules in other filamentous fungi . This hypothesis is further strengthened by an accompanying study , which also identified HAM-5 as scaffold protein of the MAK-2 pathway [55] . We hypothesize that HAM-5 and MAK-2 are co-recruited to intracellular puncta in the presumed signal receiver phase [15] , [16] and that some these puncta are subsequently targeted to the apical region of communicating cells . This process may potentially reflect the predicted auto-amplification of the incoming signal through the MAK-2 cascade or priming of the receiving cell for signal release during the next phase of communication [18] . The formation of HAM-5 puncta in the Δmak-2 mutant indicates that MAK-2 is not required for complex formation of HAM-5 . HAM-5 puncta remain stable over long time periods in Δmak-2 , and hyphae in contact frequently display distinct HAM-5 localization patterns , suggesting that the two cells are locked into two distinct signaling modes . Thus , complex dispersion and the switch from signal receiving to signal sending require MAK-2 activity . One possible mechanism for complex disassembly and termination of the signal receiving phase of the cell may involve accumulating phosphorylation of the HAM-5 scaffold through MAK-2 and/or additional unknown kinases as proposed in [55] . This hypothesis is also in line with previous work reporting that unregulated ( both reduced as well as increased ) MAK-2 activity allows reasonable tip growth of vegetative hyphae , while cell-cell communication requires the regulated on/off switch of the MAK-2 cascade [17] , [30] . In contrast to MAK-2 [17] , HAM-5 did not accumulate at the apex of non-communicating hyphae , and residual MAK-2 pathway functionality is retained in Δham-5 allowing reasonable rates of mycelial extension . These data imply a cell communication-specific function of HAM-5 for the MAK-2 cascade . In addition , previous work had identified the kinase adapter HYM-1/MO25 , which associates with multiple Ste20-related kinases in N . crassa ( this study; [17] , [47] ) and higher eukaryotes [56] as general platform that is required for MAK-2-dependent intercellular communication and for basic growth-associated functions of the MAK-2 pathway . We identified STE-20 and RAS-2 that together are critical for signal input of the MAK-2 pathway . Moreover , the AC capping protein CAP-1 is also involved in cell-cell communication . The yeast homolog of CAP-1 , Srv2p , was identified as part of a RAS-responsive AC complex in S . cerevisiae [57] , [58] . CAP-1 homologs also play a critical role in regulating actin dynamics and cell polarity in various fungi as well as higher eukaryotes [59]–[62] . Thus , the association of STE-50 with CAP-1 may link upstream components of the MAK-2 pathway with Ras/cAMP signaling and with cell morphogenesis through regulation of actin dynamics . We did not detect the second STE-20-related kinase CLA-4 as MAK-2 pathway-associated protein in our proteomics analysis , consistent with previous Y2H data that indicate no physical interaction between NRC-1 and CLA-4 [17] . Thus , CLA-4 , which was recently implicated in self-signaling in N . crassa [63] , may function as part of another module , such as the predicted BEM-1/CDC-42/RAC/CDC-24/CLA-4 complex that regulates cell polarity and potentially also chemotrophic growth [64]-[66] . Intriguingly , we also identified the Ste20-related kinase MST-1 as NRC-1- and MEK-2-interacting protein ( this study; [47] ) . Although the significance of these interactions is currently unclear , the entire MAK-2 cascade including STE-50 and HAM-5 associates with septa ( [16] , [17]; this study ) . The dynamic localization of HAM-5 at septa during intra-colony communication and its strong septal pore association may therefore indicate that MAK-2-dependent signals can originate from septa and/or that incoming signals are integrated by the MAPK pathway at these sites . This speculation is supported by accumulating evidence of central functions of septal pores as signaling hubs within the fungal colony [67]–[69] . Finally , the association of the casein kinase 2 holoenzyme and the PP2A heterotrimer with the MAK-2 cascade opens the intriguing possibility for activity regulation of the MAPK cascade . Analogous results were obtained for the mammalian casein kinase 2 , which also associates with phosphatase 2A and down-regulates the PP2A substrate MEK1 [70] , [71] . A similar , regulatory role may also be attributed to the recently defined STRIPAK complex [27] , [28] , [72] . Significantly , we did not detect the STRIPAK subunits HAM-2 and HAM-4 in our MS analysis , and thus only a sub-complex consisting of the PP2A heterotrimer and the kinase adaptor protein MOB-3 may associate with and regulate the MAK-2 cascade . MAK-2 pathway regulation may also occur through modulation of lipid composition and thus membrane dynamics , which is central for the organization and dynamic localization of multiple signal transduction pathways , including the yeast pheromone pathway [73] , and has also been implicated in self-signaling in N . crassa [74] . Yeast homologs of the N . crassa kinases YPK-1 and NRC-2 function as regulators of so-called flippase complexes [75] , [76] , which are primarily localized to the plasma membrane at sites of polarized growth , and phospholipid flipping has been shown to regulate Cdc42p signaling during polarized growth in yeast [77] . Because N . crassa ypk-1 and nrc-2 strongly phenocopy vegetative and developmental traits of MAK-2 pathway mutants [78] , [79] , the association of YPK-1 with NRC-1 and the cell communication defects detected in deletion and temperature-sensitive ypk-1 mutants will be of particular interest for dissecting oscillatory MAK-2 signaling and the chemotropic behavior of communicating cells .
Strains and oligonucleotides used in this study are listed in Tables S2 and S3 , respectively . General genetic procedures and media used in the handling of N . crassa are available through the Fungal Genetic Stock Center ( www . fgsc . net ) . Fusion proteins were ectopically expressed under the control of the ccg-1 or gpd-1 promoters at the his-3 locus . The ORFs of ste-50 , ham-5 , ste-20 and ras-2 were amplified by PCR as annotated by the N . crassa database and introduced into pMF272 , pNGFP and pJV15-2 [80]–[82] . For co-expression of tagged proteins , the pccg-1-sgfp sequence of the vector pNGFP [81] was replaced with the inversely oriented ccg-1 and gpd-1 promoters amplified from the vector pBiFC [83] to generate the vector pCCG1-pGPD1 , which allowed insertion of fusion-constructs via SgsI/SpeI ( mak-2-sgfp ) and SwaI/EcoRI ( 3xflag-nrc-1 ) amplified from the template plasmids pMF272-mak2 and pFLAG-nrc1 [17] . Resulting plasmids were transformed into his-3 and/or his-3;Δ strains . Complementation of tropic interaction , growth and developmental defects of the deletion strain was used to confirm functionality of the constructs . Co-immunoprecipitation experiments were performed as described [28] , [84] . Conditions and plasmids used for the Y2H assays are specified in [83] , [85] . Basal and stress-induced MAK-2 activity of exponentially growing N . crassa liquid cultures was determined using polyclonal rabbit α-Phospho-p44/42 MAPK ( Cell Signaling Technology , USA ) and goat α-rabbit IgG-HRP ( Santa Cruz , USA ) as primary and secondary antibodies , respectively as described [86] , [87] . Briefly , exponentially growing , liquid cultures were harvested gently by filtration using a Büchner funnel and ground in liquid nitrogen . The frozen mycelial powder was incubated in 95% ethanol at -20°C for ≥12 h , the supernatant removed after centrifugation and the pellet vacuum-dried in a SpeedVac concentrator ( Thermo Fisher Scientific , USA ) . extraction buffer ( 50 mM Tris/HCL pH 7 , 5 , 100 mM KCl , 10 mM MgCl2 , 0 . 15% NP-40 , 5 mM NaF , 1 mM PEFA , 1 mM Na3VO4 , 25 mM β-glycerophosphate , 2 mM benzamidine , 2 ng/µl pepstatin A , 10 ng/µl aprotinin , 10 ng/µl leupeptin ) was added , the samples mixed and incubated at 80°C for 5 min and the supernatant collected after centrifugation . After a second round of extraction , the supernatants pooled , subjected to another centrifugation step , and the protein concentration determined using a Nanodrop spectrophotometer ( ND-1000 , Peqlab , Germany ) . Sample volumes corresponding to 75 µg total protein per lane were subjected to SDS polyacrylamide gel electrophoresis . ≥3 biological replicates were quantified for each experiment . For quantification of MAK-2 phosphorylation levels , exposed films were scanned at a resolution of 600 dpi and densitometry was performed on the resulting tif files employing the AIDA Image Analyzer ( version 4 . 22; raytest Isotopenmessgeräte , Germany ) in transmission mode . Intensity values [arbitrary units] measured within a region of interest of fixed size containing the MAK-2 protein bands were corrected by subtraction of local background , normalized to the protein amount loaded and used for further evaluation . GFP-trap experiments , mass spectrometry and database analysis were performed as described [28] , [88] . Pulverized mycelium was mixed 1∶1 with extraction buffer and centrifuged ( 1 h , 10 , 000 rpm , Sorvall SS34 rotor ) in order to obtain crude cell extracts . Cell extracts were incubated with 2 µl GFP-trap beads ( Chromotek , Germany ) per 15 ml cell extract on a rotator for 2 h at 4°C . The beads were washed three times with IP buffer , associated proteins were recovered by boiling them in Laemmli buffer and separated by SDS polyacrylamide gel electrophoresis analysis . Peptides of in-gel trypsinated proteins were extracted from Coomassie-stained gel slices . Peptides of 5 µl sample solution were trapped and washed with 0 . 05% trifluoroacetic acid on an Acclaim PepMap 100 column ( 75 µm×2 cm , C18 , 3 µm , 100 Å , P/N164535 Thermo Scientific , USA ) at a flow rate of 4 µl/min for 12 min . Analytical peptide separation by reverse phase chromatography was performed on an Acclaim PepMap RSLC column ( 75 µm×15 cm , C18 , 3 µm , 100 Å , P/N164534 Thermo Scientific , USA ) running a gradient from 96% solvent A ( 0 . 1% formic acid ) and 4% solvent B ( acetonitrile , 0 . 1% formic acid ) to 50% solvent B within 25 min at a flow rate of 250 nl/min . Peptides eluting from the chromatographic column were on-line ionized by nano-electrospray using the Nanospray Flex Ion Source ( Thermo Scientific , USA ) and transferred into the mass spectrometer . Full scans within m/z 300–1850 were recorded by the Orbitrap-FT analyzer at a resolution of 60 , 000 at m/z 400 . Each sample was analyzed using two different fragmentation techniques applying a data-dependent top 5 experiment: collision-induced decay with multistage activation and readout in the LTQ Velos Pro linear ion trap , and higher energy collision dissociation and subsequent readout in the Orbitrap-FT analyzer . LC/MS method programming and data acquisition was performed with XCalibur 2 . 2 ( Thermo Scientific , USA ) . Orbitrap raw files were analyzed with Proteome Discoverer 1 . 3 ( Thermo Scientific , USA ) using the Mascot and Sequest search engines against the N . crasssa protein database with the following criteria: peptide mass tolerance 10 ppm , MS/MS ion mass tolerance 0 . 8 Da , and up to two missed cleavages allowed . Spinning disc confocal microscopy was performed as described [47] using an inverted Axio Observer Z1 microscope ( Zeiss , Germany ) equipped with a CSU-22 confocal scanner unit and a CCD camera ( Axiocam MRm Rev . 3 ) . ZEN Blue 2012 software ( Zeiss , Germany ) was used for image/video acquisition and image analysis . Plasma membrane was stained with FM4-64 ( 1 mg/ml-1 ) , and time-lapse imaging was performed at capture intervals of 20 s for periods up to 40 min using a C-Apochromat 63x/1 . 2 W objective . Image series were converted into movies ( * . mov ) . Tropic germling interactions of ≥100 germlings in each of 2-3 biological replicates per experiment were quantified as described [28] , [65] . In brief , strains were grown on Vogel's minimal media slants for 7 days at 26°C . Conidia were harvested with 1 to 2 ml H2O , and the conidial suspension was filtered through cheesecloth . A total number of 5×106 fresh spores were spread out on a minimal media plate , incubated at 30°C and analyzed after the indicated time points using a Zeiss Axiophot 2 microscope with a Zeiss Plan-Apochromat 63x/1 . 40 oil immersion objective . Only germlings that had produced germ-tubes of at least 2 µm length and were localized within a distance smaller than 10 µm to another germling were included in the trophic interaction analysis . | Appropriate cellular responses to external stimuli depend on the highly orchestrated activity of interconnected signaling cascades . One crucial level of control arises from the formation of discrete complexes through scaffold proteins that bind multiple components of a given pathway . Central for our understanding of these signaling platforms is the archetypical MAP kinase scaffold Ste5p , a protein that is restricted to budding yeast and close relatives . We identified HAM-5 , a protein highly conserved in filamentous ascomycete fungi , as cell–cell communication-specific scaffold protein of the Neurospora crassa MAK-2 cascade ( homologous to the budding yeast pheromone pathway ) . We also describe a network of upstream acting proteins , consisting of two Ste20-related kinases , the small G-protein RAS-2 and the adenylate cyclase capping protein CAP-1 , whose signals converge on HAM-5 . Our work has implications for the mechanistic understanding of MAP kinase scaffold proteins and their function during intercellular communication in eukaryotic microbes as well as higher eukaryotes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mitogenic",
"signaling",
"fungal",
"genetics",
"microbiology",
"fungi",
"mapk",
"signaling",
"cascades",
"protein",
"kinase",
"signaling",
"cascade",
"mycology",
"ascomycetes",
"molds",
"(fungi)",
"fungal",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"neurospora",
"crassa",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"cell",
"biology",
"cell",
"signaling",
"neurospora",
"organisms",
"signaling",
"cascades"
] | 2014 | Fungal Communication Requires the MAK-2 Pathway Elements STE-20 and RAS-2, the NRC-1 Adapter STE-50 and the MAP Kinase Scaffold HAM-5 |
Little is known about the protective role of inflammatory processes in modulating lipid metabolism in infection . Here we report an intimate link between the innate immune response to infection and regulation of the sterol metabolic network characterized by down-regulation of sterol biosynthesis by an interferon regulatory loop mechanism . In time-series experiments profiling genome-wide lipid-associated gene expression of macrophages , we show a selective and coordinated negative regulation of the complete sterol pathway upon viral infection or cytokine treatment with IFNγ or β but not TNF , IL1β , or IL6 . Quantitative analysis at the protein level of selected sterol metabolic enzymes upon infection shows a similar level of suppression . Experimental testing of sterol metabolite levels using lipidomic-based measurements shows a reduction in metabolic output . On the basis of pharmacologic and RNAi inhibition of the sterol pathway we show augmented protection against viral infection , and in combination with metabolite rescue experiments , we identify the requirement of the mevalonate-isoprenoid branch of the sterol metabolic network in the protective response upon statin or IFNβ treatment . Conditioned media experiments from infected cells support an involvement of secreted type 1 interferon ( s ) to be sufficient for reducing the sterol pathway upon infection . Moreover , we show that infection of primary macrophages containing a genetic knockout of the major type I interferon , IFNβ , leads to only a partial suppression of the sterol pathway , while genetic knockout of the receptor for all type I interferon family members , ifnar1 , or associated signaling component , tyk2 , completely abolishes the reduction of the sterol biosynthetic activity upon infection . Levels of the proteolytically cleaved nuclear forms of SREBP2 , a key transcriptional regulator of sterol biosynthesis , are reduced upon infection and IFNβ treatment at both the protein and de novo transcription level . The reduction in srebf2 gene transcription upon infection and IFN treatment is also found to be strictly dependent on ifnar1 . Altogether these results show that type 1 IFN signaling is both necessary and sufficient for reducing the sterol metabolic network activity upon infection , thereby linking the regulation of the sterol pathway with interferon anti-viral defense responses . These findings bring a new link between sterol metabolism and interferon antiviral response and support the idea of using host metabolic modifiers of innate immunity as a potential antiviral strategy .
Sterols and fatty acids are common intermediary metabolites that play key roles in many biological pathways involved in inflammatory diseases such as atherosclerosis and chronic heart disease [1]–[4] . Significantly , mounting evidence shows a connection between innate immune signaling processes and the regulation of sterol and fatty acid metabolism [5]–[8] . Specifically , cholesterol and its metabolites have been shown to alter inflammatory mediator behavior [9]–[11] , and conversely , innate immune signaling has been shown to modulate the dynamics of cholesterol transport , storage , and excretion [12]–[15] . Recent studies have also begun to show that the perturbation of lipid metabolism in a range of virally infected cells is a hallmark of cellular changes associated with infection . For instance , studies analyzing the consequences of human cytomegalovirus ( CMV ) infection have shown that increases in the flux of the fatty acid biosynthesis pathway are essential for optimal viral growth in fibroblasts [16] . Further , Hepatitis C virus ( HCV ) has been shown to co-opt the prenylation pathway to promote the efficient replication of its genome [17]–[19] . More generally , a number of other viruses , notably Measles , HIV , West Nile virus , and Dengue virus , also have the ability to change cholesterol pathway gene expression in a variety of cellular systems [20]–[24] . Whether the effects of virus infection on the cholesterol pathway are directly mediated by the pathogen or indirectly host-mediated mechanisms is not known . From a therapeutic perspective , studies have also shown that the pharmacological disruption of the cholesterol metabolism by statins and other metabolic modifiers can result in the inhibition of viral replication [25]–[32] . It is well documented that the cross-talk between immune programs of macrophage activation and lipid homeostasis plays a central part in chronic inflammatory diseases [33] , [34] . In particular an anti-atherosclerosis transcriptional axis of PPARγ regulating a pathway of cholesterol efflux by inducing ABCA1 expression and cholesterol removal from macrophages , via a transcriptional cascade mediated by activated LXRα , has been reported [35] . Significantly , cellular metabolic , signaling , and regulatory pathways also play a critical “collaborative” role in modulating immune responses to infection [36] . In this context , Toll-like pathogen recognition receptors , crucial to the initiation of innate immune signaling , have recently been shown to regulate the expression of key lipid-associated genes following bacterial infection . This occurs due to microbial ligand activation of the IRF3 pathway , which blocks the induction of LXR target genes such as ABCA1 and inhibits cholesterol efflux from macrophages in an interferon independent manner [5] . In this context , LXRα−/− mice are more susceptible to bacterial infection [37] , further emphasizing the importance of this pathway in the innate immune response . From a viral perspective , an interferon-inducible protein “viperin” is known to inhibit influenza A virus and HCV by disrupting the formation of cholesterol-enriched lipid rafts , which act as attachment sites for viral production [38] , [39] . Significantly , despite increasing numbers of studies in this area , the question remains as to whether the immune regulation of lipid pathways can also serve a role as part of a protective anti-viral response . Indeed , in the context of host protection pathways , it is not known whether a central immune regulatory mechanism involving interferon response is directly or indirectly required in modulating lipid metabolism in infection . We are interested in elucidating the relationship between transcriptional networks and immune regulatory pathways and host-cell dependency mechanisms of pathogens , especially viruses , as identifying host dependency mechanisms at the pathway level provides a new molecular systems-level approach for understanding viral pathogenesis , which can be harnessed as an anti-infective strategy [40]–[42] . For many years , studies of virus-host interactions , in particular for large DNA viruses , have proven invaluable in the characterization of host cell molecular pathways and their connectivity to the inflammatory response . Murine cytomegalovirus ( mCMV ) , which has a large double-stranded DNA genome , represents one of the few model organisms studied in its natural host and has both biological and clinical relevance to human CMV disease [43] . In this study , we have sought to apply a systems-level approach , bringing together functional genomics , lipidomics , and biochemical experimentation , to understand the interplay between sterol pathway down-regulation and the innate immune response to mCMV infection . Our investigations reveal a previously undisclosed dependency role for down-regulation of the sterol metabolic network , which is integral to the protective immune response requiring a type 1 interferon receptor regulatory loop mechanism .
As a first step , an integrative approach combining bioinformatics tools and a time-series analysis of gene expression changes was applied to mCMV-infected or interferon ( IFN ) γ-activated primary bone-marrow-derived macrophages ( BMDM ) . These primary BMDM cultures represent a physiologically relevant cell system for the combined analysis of infection , inflammation , and lipogenesis [44]–[46] . In the following experiments , infected or IFNγ-treated BMDM RNA was harvested every 30 min up to 12 h post-challenge for microarray gene expression profiling . In this study , analysis of expression data was exclusively restricted to lipogenic-associated genes . For this purpose , a combination of literature and data-mining identified over one thousand genes with published direct or indirect functions relating to cellular lipid metabolism , regulation , and synthesis ( Text S1 ) . When this resource was used to interrogate a subset of our time-series data which passed a stringent filtering threshold ( p<10−6 ) , 89% of lipogenic-associated genes were detected , of which 12% were significantly regulated ( 113/958 ) upon IFNγ treatment and 23% were significantly altered in their expression ( 195/958 ) after mCMV infection . This represented a significant and highly selective lipogenic response ( Figure S1 ) with altered genes showing a high degree of overlap between infection and IFNγ activation ( Table S1 ) . Notably , clear differences in the specific class of lipogenic genes in up- and down-regulated groups were observed . Of the IFNγ down-regulated transcripts , a significant proportion ( 14/35 , 40% ) were related to the sterol pathway , while fatty acid pathways were pre-eminent ( 6/35 , 17% ) in the up-regulated gene group ( Figure S1C ) . A statistical evaluation investigating pathway over-representation indicated a highly pathway-specific response including previously known pathways for inositol ( Table S2E–F ) [47] perturbed by mCMV infection . Significantly , however , the most pronounced pathway changes in the down-regulated genes common to both stimuli were associated with sterol lipid metabolism ( Table S2E and Figure 1A ) , which exhibited a gradual , temporal decline in expression from 6 h post-infection ( hpi ) onwards ( Figure 1B ) . Additional microarray experiments to further explore this observation revealed a further reduction in sterol pathway gene expression observed at 24 hpi ( unpublished data ) . It is worth noting , however , that the observed level of reduction in expression for any particular transcript was relatively modest ( ranging from 1 . 3- to 5-fold for infection and 1 . 3- to 3-fold for IFNγ treatment over a 24 h time frame ) . To independently validate the microarray data described above , Q-RT-PCR analyses of five independent experiments were performed for both infection and IFNγ treatment . In agreement , we find that Q-RT-PCR analysis of selected members of the pathway—Hmgcs1 , Hmgcr , IdI1 , and Sqle—shows a statistically significant but quantitatively modest reduction in expression ( Figure 1C and 1D ) . Notably , a similar quantitative decrease is also exhibited at the protein level for HMGCS1 , HMGCR , and SQLE ( Figure 2A ) . Since the alterations in expression of the cholesterol-related genes were consistent but of relatively small magnitude , we considered whether these perturbations represented either non-specific “noise” generated during the pro-inflammatory stimulation of a macrophage or a more specific response to a particular challenge . To test whether alternative pro-inflammatory mediators could also lead to the modulation of the sterol pathway genes , macrophage cultures were treated with a range of doses of the following inflammatory cytokines: IL1β , TNF , IL6 , and IFNβ . Hmgcs1 , Hmgcr , Idi1 , and Sqle gene expression changes were then analyzed by Q-RT-PCR ( Figure 1E–H ) . Of the cytokines tested , only IFNβ elicited the down-regulation of sterol pathway gene expression in primary macrophage cultures ( Figure 1E ) . In summary , these data indicate a highly specific response of macrophages through a coordinated negative regulation of multiple sterol pathway members upon viral infection or treatment with IFNγ or β but not IL1β , TNF , or IL6 . Once again , these effects are quantitatively “modest” but statistically significant . We next sought to explore how multiple small reductions in enzyme levels impact upon the biosynthetic activity of the pathway by measuring the steady-state metabolic output of the pathway . For these experiments , free intra-cellular cholesterol level , as a metabolic end product of the sterol pathway , was determined using an enzymatic method on infected macrophages ( Figure 2B ) . We observe a significant decrease in cholesterol metabolite levels 24 hpi . Similar results were also observed with infection of NIH/3T3 cells ( Figure 2C ) , indicating that the effect is not macrophage specific . It is possible that the experimentally observed drop in sterol lipid levels could be due to a non-specific and generalized response to infection , although from the microarray analysis of the lipidomic associated genes we clearly observe highly specific lipogenic responses rather than a broad response to infection ( Figures S1 and S2 ) . To further determine whether the down-regulation of sterol biosynthesis is specific between mCMV infection and select lipogenesis pathways , total cell extracts were analyzed by electrospray ionization as well as atmosphere chemical ionization mass spectrometry ( see Materials and Methods ) . These lipidomic approaches allow quantification of the major membrane lipid classes ( such as glycerophospholipids and sterols ) as well as individual molecular lipid species at high sensitivity . Overall , we find no coordinated or substantial differences in the overall levels of major glycerophospholipids ( phosphatidylcholine , phosphatidylserine , and phosphatidylethanolamine ) during infection with CMV , although a small number of the individual species in the phosphatidylcholine and phosphatidylserine group are affected ( Figure S3A–C ) . In marked contrast , levels of free cholesterol , as well as its immediate precursor , zymosterol , 14-demethyl-lanosterol , and 7-dehydrocholesterol , were strongly reduced at 24 hpi ( 2–3-fold ) and 48 hpi ( 4–6-fold ) ( Figure S2A–D ) . These results further support a specific alteration of sterol biosynthesis upon infection . Furthermore , the reduced free cholesterol levels are also developed in a dose-dependent manner by treatment with IFNβ and γ but not IL1β , IL6 , or TNF ( Figure 2D ) . Altogether , we conclude that the effect of the coordinated down-regulation is to reduce metabolic output of the sterol pathway . To assess whether the sterol biosynthesis pathway plays a pro- or anti-viral role in regulating mCMV replication , we exploited the pharmacologic compound “simvastatin , ” a potent and selective inhibitor of HMGCR [48] . Inhibition of HMGCR is known to result in a reduction of the metabolic intermediate mevalonate ( Figure 3 ) and an accompanying drop in cholesterol synthesis by the cell [49] . The treatment of cells with simvastatin resulted in a dose-dependent inhibition of mCMV plaque formation ( unpublished data ) and in live cell replication assays ( Figure 4A ) with an IC50 of 2 µM that is comparable to the “gold standard” anti-viral Gancyclovir ( Figure 4A ) in the murine model system . Notably , the observed inhibitory effect of simvastatin occurred below a level at which non-specific toxic effects to cells were observed ( 15 µM ) ( Figure S7 ) . These experiments pointed to a potential protective anti-viral role via a targeted disruption of the sterol pathway and raised the question of whether pharmacologic treatment in vivo also develops an inhibitory effect . To investigate whether simvastatin could play an anti-infective role in vivo , mice were administered with an established pre-clinical pharmacologic dose of simvastatin or vehicle alone and infected by intra-peritoneal inoculation with mCMV . Viral titres were then determined in a variety of organs at day 4 post-inoculation . Markedly , viral titres are reduced by over one order of magnitude in multiple organs following treatment with simvastatin ( Figure 4B ) . To determine the extent of the overlap between the sterol biosynthesis pathway and anti-viral activity , we employed a series of metabolite rescue and interference RNA knock-down experiments . In these experiments we observed that simvastatin anti-viral activity could be completely reversed by the addition of mevalonate to cells in culture ( Figure 5A ) . This showed that the anti-viral mechanism was due to an inhibition of HMGCoA reductase . While this result supports the requirement of the mevalonate arm of the sterol pathway , it does not necessarily implicate cholesterol as being responsible for the anti-viral activity . Notably , feeding macrophages with a cell permeable form of cholesterol or squalene failed to reverse the inhibitory activity ( Figures 3 and 5A ) , indicating that the anti-viral effect is unlikely to be cholesterol mediated and thus unrelated to any regulatory sterols or to the structural requirements of virus replication associated with lipid droplets . The addition of cell permeable farnesol also did not rescue the inhibitory activity of simvastatin , while conversely the addition of geranylgeraniol fully rescued the anti-viral activity ( Figures 3 and 5A ) . These experiments show the specificity of the metabolic requirement for anti-viral activity and highlight a possible role for the mevalonate-isoprenoid arm of the sterol pathway in protection against mCMV infection ( Figure 3 ) . While the rescue of statin inhibition of viral growth by mevalonate and geranylgeraniol strongly indicates the involvement of the proximal arm of the sterol pathway , it is still conceivable that other mechanisms of action unrelated to the capacity to inhibit biosynthesis may be responsible for the effect on virus replication . For this reason and to additionally test the specific requirement of members of the sterol biosynthesis pathway for viral growth control , siRNA knock-down experiments were performed . For these experiments , Hmgcs1 and Hmgcr were first targeted in the pathway using low concentrations of siRNA to avoid non-specific interferon responses , including a series of non-targeting siRNA for non-targeting effects . Knock-down of these genes ( Figure S5 ) resulted in a specific and significant decrease in the optimal rate and end-point yield of viral replication ( Figures 5B ) . To further dissect the specific role of pathway members in mediating the anti-viral response , additional siRNA inhibition studies were conducted involving targeted genes distal to the mevalonate-prenylation branch of the sterol biosynthesis pathway . In these experiments , Fdft1 , Sqle , and Dhcr7 were targeted . Figure 5B clearly shows that targeting these members of the pathway fails to inhibit and even positively influences viral growth , a result that is consistent with the above described metabolite rescue experiments . To further investigate and to independently assess the specificity of the prenylation branch of the pathway , additional siRNA knock-down experiments were performed targeting farnesyl diphosphate synthase ( Fdps ) , an enzyme essential for isoprenoid biosynthesis , and all three prenyltransferases ( these are farnesyltransferase , geranylgeranyltransferase type I , and Rab geranylgeranyltransferase type II enzymes ) . In these experiments knock-down of Hmgcr and Dhcr7 and viral ORFs ( M54 and M86 ) are used as controls and developed the expected knock-down profile ( Figure 5C ) . Notably , significant inhibition of viral replication is observed for knock-down of Fdps . In the case of the downstream prenyltransferases , reduced viral replication is observed with siRNA targeting Rabggtb specific for geranylgeranyltransferase type II enzyme , but not Pggt1b or Fntb specific for gernylgeranyltransferase type I and farsenyltransferase , respectively ( Figure 5C ) . These experiments indicate specificity of targeting the isoprenoid pathway but will require further functional validation work . Overall , these findings show that inhibition of viral growth is not due to cholesterol deprivation , but rather a part of the pathway involving a proximal mevalonate-prenylation step . This raises the notion of whether depletion of geranylgeraniol may be one potential mode for interferon to inhibit viral replication . In this scenario we might expect that feeding cells with geranylgeraniol upon interferon treatment would counter , in part , the anti-viral effect . To determine the effect of interferon on viral replication , in the absence and presence of geranylgeraniol , we performed a metabolite rescue experiment in the presence of increasing units of IFNβ . Figure 5D shows that the anti-viral effect of 1 and 5 U/ml of IFNβ is dramatically reduced in the presence of geranylgeraniol ( at both 15 and 150 µM ) , while at a more potent level of IFNβ ( at 25 U/ml ) approximately 70% and 25% of the anti-viral activity remains with 15 and150 µM GGOH , respectively . Taken together , these results support a role of the mevalonate-isoprenoid arm of the sterol pathway for optimal mCMV replication and highlight the potential role for down-regulating this pathway in protecting the host from viral infection . These findings also suggest that sterol biosynthesis regulation acts as a marker for antiviral activity . We next sought to investigate whether specific viral or cellular modes of action might be responsible for the reduction in sterol biosynthesis upon infection . First , it is possible that the effects monitored in our experimental system are specific to mCMV . To test whether the down-regulation of sterol pathway gene expression is a more general effect rather than specific to mCMV , primary macrophages ( BMDM ) were infected with a number of different viruses and harvested for gene expression analysis . Figure 6A shows the expression profile of the sterol pathway and other pathways for infection ( innate immune activation pathways ) by an enveloped DNA virus , herpes simplex ( HSV1 ) ; an RNA virus , semliki forest virus ( SFV ) ; cytoplasmic DNA virus , vaccinia virus ( VV ) ; and non-enveloped nuclear DNA virus , adenovirus . All the viruses tested show a specific and coordinate decrease in gene expression for members of the sterol biosynthesis pathway . In the case of mCMV , it is worth noting that the reduction in gene expression occurs approximately 6 hpi ( e . g . , see Figure 1 ) . Consequently , it is possible that a viral early or late gene product may be required for the effect . To test this possibility , we used a replication and early/late gene defective mCMV virus ( mCMVdie3 in [50] ) . The mCMVdie3 strain is capable of infecting cells at levels equivalent to wild-type virus but is incapable of expressing its genome downstream of a rather restricted immediate-early phase . The results of these experiments are shown in Figure 6B , in which mCMVdie3 potently develops an equivalent level of down-regulation of sterol genes as the parental wild-type and revertant viruses , respectively . It is well established for many viruses , including mCMV , that infection leads to the induced expression of type 1 interferon and pro-inflammatory cytokines . Two signaling cascades—a virus-induced interferon-producing signal and an interferon receptor-mediated secondary signal—regulate the interferon system . The first is initiated by the detection of viral components by host recognition receptors ( PRRs ) and leads to the activation of transcription factors—NFkB , ATF2/c-Jun , IRF3 , and IRF7—that activate IFNα and β genes . The expressed interferons then transmit a secondary autocrine or paracrine signal through interactions with type I receptors that activate the JAK-STAT pathway . In this context , the above studies with the combined observation that interferon treatment and the cell response to infection are equally capable of causing a down-regulation of the sterol metabolic pathway raise the question of whether infection-mediated regulation might result from an interferon regulated loop . In support of this notion we find that low multiplicities of infection still exhibit a significantly reduced level of free cholesterol ( Figure S6 ) and that conditioned media from infected macrophages 8 hpi ( prior to release of any new viral particles ) are sufficient to down-regulate the sterol biosynthesis pathway genes in uninfected control cultures ( Figure 6C ) . On the basis of temporal expression , causal inference of candidate effectors can be tested . A search of cytokine profiles suggested a strong correlation following IFNβ synthesis , further raising the hypothesis for a potential interferon regulatory loop mechanism that is responsible for modulating sterol biosynthesis . First , we investigated directly whether IFNβ is responsible by infecting BMDM from ifnβ−/− mice and examining gene expression for representative members of the sterol pathway . Figure 7A shows that following the genetic ablation of IFNβ , there is still statistically significant sterol gene expression reduction but that there is a partial loss in the degree of reduction indicating that IFNβ is not absolutely necessary . It is possible that other type I IFN members may compensate for the lack of IFNβ . The redundancy among the various type I interferons can be directly evaluated through genetic knockout of their shared receptor , IFNAR1 . For this reason we next investigated whether the sterol response to infection is dependent on the type I interferon receptor . To this end , primary macrophages derived from IFNAR1−/− mice were challenged with mCMV or IFNβ , and the sterol biosynthesis gene expression and free cholesterol levels were analyzed . As shown in Figure 7B–D , the lack of interferon type I receptor abolished the ability of macrophages to reduce both sterol biosynthesis gene expression and cholesterol yield upon either infection with mCMV or treatment with IFNβ . We conclude from these experiments that a type I interferon-dependent innate immune response stringently regulates the metabolic alteration of the sterol biosynthesis network observed upon infection . Type 1 interferon has an important role in the control of mCMV replication , and the tyrosine kinase 2 ( Tyk 2 ) signaling component is absolutely essential for the type I defense against mCMV infection . Notably , the lack of Tyk2 is known to selectively impair the transcription of only a subset of virally induced IFNAR1 responsive genes [51] . Since this occurs at the promoter-transcriptional level , we first asked whether the down-regulation of the sterol pathway in response to infection also occurs at the level of gene transcription . For this purpose and to directly measure the level of de novo transcription of members of the sterol pathway , we exploited a recently established labeling protocol for the isolation and analysis of newly transcribed RNA [52] . In these experiments , macrophages were infected with mCMV in the presence of 4-thiouridine , for 30 min at 6 hpi , allowing efficient labeling of nascent RNA for isolation and interrogation by microarray analysis . Figure 8A shows that infection by mCMV results in the anticipated reduced level of newly transcribed RNA of the sterol biosynthetic pathway genes . Next we sought to test whether the Tyk2 receptor-signaling component is required for the type I interferon-dependent down-regulation of the sterol pathway . For these experiments we used tyk2−/− macrophages and observe an almost complete abrogation of the transcriptional down-regulation by mCMV infection ( Figure 8B ) . These results demonstrate a requirement for Tyk2 in the mCMV-mediated gene down-regulation of the sterol biosynthesis pathway and suggest a novel role of interferon type I receptor signaling as a transcriptional modifier of the host's metabolic response to infection . The above studies strongly point to a transcriptional mechanism in down-regulating the sterol pathway upon infection . The sterol regulatory binding protein 2 ( SREBP2 ) is the principal transcription factor involved in coordinating the regulation of the sterol biosynthesis pathway [53] . SREBP2 is synthesized as a precursor and anchored in the endoplasmic reticulum membrane and through limited proteolysis is activated to generate mature forms that can enter the nucleus and interact with multiple sterol pathway genes to coordinate their expression . Hence , in order to gain further insight into the potential mechanism for participating in the transcriptional down-regulation of the sterol pathway , we investigated in the first instance the protein levels of activated cleaved forms of SREBP2 upon infection and interferon treatment . Accordingly , we next performed Western blot experiments to determine levels of mature form of SREBP2 . In these experiments , infection of macrophages with mCMV at 24 hpi developed a significant decrease in the nuclear form ( Figure 9A ) . Furthermore , treatment of macrophages with either IFNβ or IFNγ clearly exhibits a decrease in SREBP2 levels ( Figure 9A ) . We next sought to examine whether this is also seen at the level of transcription . In experiments measuring de novo RNA synthesis , we observed a specific transcriptional reduction from the Srebf2 gene upon infection while increased levels of transcription are seen for interferon-associated transcription factor Stat1 ( Figure 9B ) , indicating a selective transcriptional basis for the reduced levels of expression . Markedly , the reduction in RNA levels upon infection was completely reversed upon genetic ablation of the ifnar1 gene ( Figure 9C ) . Altogether these results demonstrate a coordinate reduction in SREBP2 at both the protein and RNA expression level upon infection , which is tightly dependent on activation of the type 1 interferon receptor .
Our results are consistent with a model involving a two-step interferon response for modulating endogenous sterol pathway activity upon infection . Figure 10 illustrates the two signaling cascades , a virus-induced interferon-producing signal and an interferon receptor-mediated secondary signal . The first is initiated by the detection of virion proteins and nucleic acids by host recognition receptors with the result of the infected cell producing type I interferon . As part of the second step all type I interferons bind to one common receptor ( IFNAR1 ) . The IFN-α/β receptor ( IFNAR1 ) signals through the JAK/STAT pathway by phosphorylation of the Janus kinase ( JAK ) 1 , tyrosine kinase ( Tyk ) 2 , and signal transducer and activator of transcription ( STAT ) 1 and STAT2 , which subsequently modulates a diverse array of genes . In the case of mCMV the first step has been extensively investigated and shown to involve TLR2 , TLR3 , and TLR9 recognition receptors [54] , [55] , whose activation leads to the induction of transcription factors , NFkB , ATF2/c-Jun , and IRF3 that directly activate IFNα and β genes . Interestingly , previous studies [5] , [37] have shown that microbial activation of TLR3 or TLR4 inhibits by an as-yet unknown mechanism LXR target genes such as ABCA1 , resulting in the inhibition of cholesterol efflux from macrophages . This is reported to occur in a type I interferon-independent manner [5] . Similar to microbial-mediated TLR activation of IRF3 , many viruses including mCMV potently induce IRF3 and may also have the potential to inhibit LXR functions . Despite recent progress in the definition of links between intracellular cholesterol homeostasis and innate immunity , little is known regarding the influence of interferon-regulated signaling on this phenomenon . In the present study , we demonstrate that transcriptional regulation of the cellular sterol biosynthesis pathway upon infection has an impact on viral replication and depends on an interferon-regulated loop involving type 1 interferon signaling . Specifically , we show that infection of cells by a wide range of viruses or direct interferon stimulation is accompanied by the down-regulation of sterol biosynthesis as a result of reducing the rate of sterol gene transcription . In the context of ligand-activation of the type I receptor , we also demonstrate that this requires the Tyk2 signaling component ( Figure 10 ) . In the context of type I interferon genes induced upon infection , it is worth noting that viperin , a type I interferon-regulated gene , is involved in cellular defense against a number of viruses and functions to disrupt cholesterol-rich lipid rafts that are used as viral production sites in the cell [34] , [38] , [56] . In addition , an intracellular interaction of viperin with Fdps , an enzyme essential for isoprenoid biosynthesis ( Figure 3 ) , has been reported to lower , by a small extent , the activity of the enzyme [38] . It is not known whether targeting Fdps enzyme activity alone is an effective anti-viral mechanism , although the RNAi targeting results of Figure 5 ( panel C ) suggest that this may be a plausible mechanism ( Figure 5C ) . However , it is more likely that a combination of interferon-mediated transcriptional down-regulation of the sterol biosynthesis genes and the potential enzymatic protein modification at the isoprenoid branch point represents a concerted anti-viral host defense mechanism . From a transcriptional perspective , the sterol biosynthesis pathway genes are co-ordinately controlled by the sterol regulatory element binding protein 2 transcription factor ( SREBP2 ) . Significantly , we find in our system that the overall abundance of the mature protein ( the proteolytically cleaved active form ) and the rate of gene transcription of its gene are significantly reduced upon infection or interferon treatment . Significantly , both are strictly dependent on the presence and activation of the type 1 interferon receptor Ifnar1 . These findings suggest that a possible mechanism for the coordinate down-regulation of sterol biosynthesis is by interferon regulation of Srebf2 . Interestingly and consistent with the possibility of interferon regulating Srebf2 , chemical inhibition of SREBP2 has been shown to inhibit HCV replicon activity [32] . This would also support the view of implicating negative feedback on SREBP-2 via oxysterol metabolites . Further studies are required to elucidate more precisely the mechanism or mechanisms by which interferon mediates down-regulation of the sterol biosynthesis pathway . Whatever the mechanism , the IFN-dependent coupling of the mevalonate-sterol metabolic network and anti-viral activity represents a previously unrecognized mechanism in the regulation of protective immunity . From an immune response and metabolic/pharmacological perspective , modulating cholesterol biosynthesis via small , coordinate transcriptional changes offers advantages and disadvantages over single enzyme control . At the homeostatic level , coordinate control of a metabolic pathway could potentially increase the robustness of modulation; the redundant rate-limiting interactions , downstream of the true rate-limiting interaction , can protect the pathway from surges in the levels of downstream metabolites . Coordinate control also increases the specificity of the pathway modulation as a small reduction of the enzyme level in an interaction ensures that the level of the interacting metabolite need not drop as far to affect a reduction in flux . This has the advantage of potentially lessening the impact on other branched or cross-linked pathways that use the same metabolites and thus provides a high degree of pathway specificity . Several viruses including human CMV have been reported to be sensitive to statin administration [25] , [26] , [28]–[31] . Although the mechanism of action of most is not known , it has in some cases been correlated with a lower abundance of cholesterol in lipid rafts of cell membranes . A recognized potential complicating factor of using statins to specifically reduce cholesterol levels is that suppression of the proximal mevalonate arm also perturbs the synthesis of branch derivatives such as geranylgeraniol and farnesol involved in the protein farnesylation and prenylation pathways . In the case of HCV , the mechanisms of the inhibitory effects of the statins have been examined extensively and have been shown to relate to the prenylation of a host protein ( FLB2 ) essential for viral replication [18] , [57] . Recently a combination chemical screening study has been conducted to explore how the sterol and protein prenylation pathways work together to affect HCV in a replicon assay [32] . In agreement with those studies we also find reduced mCMV growth in siRNA knock-down experiments targeting enzymes in the isoprenod biosynthesis pathway . These studies indicate the importance of the geranylgeranylation to viral replication . Although , it is worth noting that the isoprenoid biosynthesis pathway is highly complicated with multiple branch points involving redundant enzymatic steps , sharing of subunits , and competing reactions . In our current study , we uncoupled the cholesterol synthesis pathway from non-steroidal modifications through targeted metabolic rescue and siRNA knock-down studies of mCMV and reveal an absolute requirement for the prenylation branch of the sterol pathway for mediating anti-viral effects . As further indicated from computational modeling work ( unpublished data ) , targeting HMGCR is likely to have a broad range of non-specific effects on various efferent branch points of the pathway and thus may well not be ideal for anti-infective therapy . In addition , statins are also known to have a range of immune-modulatory activities by mechanisms yet to be fully characterized . In this context , it is worth noting that the activity of the type I interferons , especially IFNβ , have considerable overlap with many of the immune-related activities of statins [58] . Moreover , it is especially noteworthy that IFNβ treatment in patients has also been reported to have decreased plasma cholesterol levels [59] , [60] . Since our studies uncover a molecular dependency of type 1 signaling , including a Tyk2 signaling component , this may provide an entirely new therapeutic pathway for lowering cholesterol . Moreover , our findings may have important implications for the development of broadly active new adjuvant strategies ( e . g . , the use of inhibitors of SREBP2 activity ) to existing anti-infective therapies ( e . g . , antiviral drugs such as ganciclovir ) . On this basis we posit the principal of using metabolic modifiers , i . e . drugs that target metabolic pathways , of protective innate immunity as holding future promise for developing host-directed anti-viral therapies . Overall , this study supports the original concept [40] , [41] of selectively targeting host pathways as an efficacious anti-infective strategy .
Microarray analysis of the time course experiments of infected and interferon treated macrophages were conducted using Agilent microarray platform and a detailed description of the experimental set up; statistical and bioinformatics analysis is in the Supporting Information section . All other microarray studies were conducted using Affymetrix ( Mouse Genome 430 ) microarray platform . Data from hybridized Affymetrix microarrays were acquired using proprietary Affymetrix platform scanners and GCOS software ( Affymetrix ) . Processed CEL files were imported into Partek Genomics SuiteTM ( MO , USA ) , then background corrected , quantile normalized , and probe-set summarized using the RMA algorithm [61] , [62] . A non-specific filter was applied to remove genes that were not expressed on any of the samples across the experiment . Microarray signals were then per-gene normalized to the average of the three mock samples ( which was set to a value of 1 ) for visualization purposes in the heat map for Figure 6 . In the case of de novo RNA expression , analysis was performed using the Affymetrix Mouse Gene 1 . 0 ST arrays , consisting of a total of eight chips and three experiment factors: time ( 60–90 min , 360–390 min ) , genetic background ( Tyk2KO , WT ) , and treatment ( mock , mCMV ) . Data from hybridized chips were acquired using GCOS software ( Affymetrix ) . Prior to further processing and analysis with the R statistical programming environment , Affymetrix Power Tools ( APT , Affymetrix ) were used to summaries and annotate chip data to gene level . After initial quality control assessment , data were background-corrected , quantile normalized , and probe-set summarized using the RMA algorithm . Wild type C57BL/6 and BALBc were from the Biomedical Research Resources , Little France , University of Edinburgh . IFNβ−/− and Tyk2−/− mice were from the Institute of Animal Breeding and Genetics Veterinary University of Vienna . BMDM were derived from monocytes obtained from femurs of male mice aged 10 to 12 wk . Cells were grown in DMEM-F12 media supplemented with 10% L929 cell-conditioned medium as a source of macrophage colony-stimulating factor ( M-CSF ) for 7 d as described [63] . Characterization of BMDM was performed by standard flow cytometry , evaluating the presence of the F4/80 marker and CD11b surface protein . In average of all experiments more than 93% of cells possessed both proteins . The mouse CMV C3X strain , generated from the recombinant C3X bacterial artificial chromosome clone and originally derived for the Smith strain of mCMV [64] , was propagated in NIH 3T3 cells , and titers were determined by standard plaque assay on MEFp53−/− . For live cell assay , NIH/3T3s were infected with a recombinant mCMV expressing the green fluorescent protein ( GFP ) marker inserted in front of the ie2 gene ( pSM3fr-rev , called mCMV-GFP in this study [65] ) . Viral growth curves comparing wild type and GFP virus were assessed by standard plaque assay , and the results showed no differences between the growth curve of the two viruses ( unpublished data ) . To establish the role of viral gene expression in the regulation of sterol genes , the mCMVdie3 strain was used [50] . For the microarray experiment , Semliki Forest Virus ( SFV , MOI of 10 ) , Herpes simplex virus type 1 ( HSV1 , MOI of 1 ) , Vaccinia virus ( VV , MOI of 1 ) , and Adenovirus ( Ad , MOI of 100 ) were used to infect BMDM for 1 h in DMEM:F12 3% FCS , 10% L929 , and 100 U of penicillin/streptomycin per ml . BMDM and NIH/3T3 were infected with the different viruses at an MOI of 1 , unless specified . For BMDM , viral stock was diluted in DMEM:F12 3% FCS , 10% L929 , and 100 U of penicillin/streptomycin per ml , and after 1 h adsorption , cells were washed in PBS and incubated in fresh DMEM:F12 10% FCS , 10% L929 , and 100 U of penicillin/streptomycin per ml . For NIH/3T3 viral stock was diluted in DMEM 3% CS and 100 U of penicillin/streptomycin per ml , and after 1 h adsorption , cells were washed in PBS and incubated in fresh DMEM:F12 10% CS and 100 U of penicillin/streptomycin per ml . SFV ( MOI of 10 ) , HSV1 ( MOI = 1 ) , VV ( MOI of 1 ) , and Ad ( MOI of 100 ) were used to infect BMDM for 1 h in DMEM:F12 3% FCS , 10% L929 , and 100 U of penicillin/streptomycin per ml . IFNγ ( Boehringer Manheim Corp ) , IFNβ , IL6 , TNF , and IL1β ( Biosource International , USA ) stock were dissolved in PBS supplemented with 0 . 2% BSA and diluted in fresh media just prior to the experiment . The effect of cytokine treatment on cell viability was tested for each concentration used in the experiment and did not show any alteration of viability . For the pharmacological experiment , 25 mg of simvastatin ( Sigma-Aldrich ) was activated by hydrolysis of the lactone by adding 1 ml of 0 . 1 N NaOH , 100% ethanol . After heating at 50°C for 2 h , the solution was neutralized with HCl to a pH of ≈7 . 2 and sterilized by filtration through a 0 . 2 µm filter . The stock solution was diluted to the appropriate concentration in sterile PBS and the solution was aliquoted , stored at −20°C , and used within a month of activation . Mevalonate and water soluble cholesterol ( Sigma-Aldrich , Germany ) was resuspended in media to the appropriate concentration and sterilized by filtration through a 0 . 2 µm filter . Geranylgeraniol and farnesol squalene ( Sigma-Aldrich , Germany ) stocks were dissolved in DMSO and sterilized by filtration through a 0 . 2 µm filter . The stock solutions were dissolved in media at the appropriate concentration just prior to the experiment . The final concentration of DMSO in media did not exceed 0 . 1% . Effects of sterol intermediates treatment on the cell were tested for each concentration used in the experiment and did not show any alteration of viability . Gancyclovir ( Cymevene , Hoffman-La Roche , UK ) was resuspended in saline solution and sterilized by filtration through a 0 . 2 µm filter . Gancyclovir was then diluted in media , to the indicated concentration . Taqman Primer probe sets were purchased from Applied Biosystems , Warrington , UK ( Assay ID: Hmgcs1: Mm00524111-m1; Hmgcr: Mm01282499-m1; Idi1: Mm00836417-g1; Sqle: Mm00436772-A1 ) . For each sample QRT-PCR was performed in 20 µl volumes using MicroAmp Optical 96-well reaction plates and MicroAmp Optical Caps ( Applied Biosystems ) . Two microliters of diluted RNA samples ( ≈100 ng of RNA ) were added to 10 µl of 2× PCR master mix , 1 µl of a Taqman primer/probe set ( Applied Biosystems , CA ) for the gene of interest at the recommended concentration , 0 . 25 µl of Superscript III ( Applied Biosystems , CA ) , and 6 . 25 µl of double-distilled H20 . After an initial incubation at 50°C for 30 s to activate the RNA polymerase , samples were then subject to 40 cycles under Taqman standard conditions ( combined annealing and primer extension phase at 60°C for 1 min and a short denaturation at 72°C for 30 s ) . Stratagene MXPro software was then used to analyze the data . Threshold determinations were automatically performed by the instrument for each reaction . The CT values were exported into Microsoft Excel and relative quantification of marker gene mRNA expression was calculated with the comparative CT method [66] . BMDM cells were washed with PBS and resuspended in whole-cell lysis buffer ( 50 mM Tris-HCl , pH 7 . 5 , 100 mM NaCl , 1% NP40 , protease inhibitors , and phosphatase inhibitors ) , and cell lysates were centrifuged at 4°C for 10 min and the collected supernatants were stored at −20°C . Protein concentration was measured by Pierce BCA assay ( Thermo Scientific ) . For Western blotting , proteins were separated by SDS-PAGE , transferred to Immobilon-FL membranes ( Millipore ) , and probed with goat anti-HMGCR ( Santa Cruz , sc-27578 , 1∶500 ) , goat anti-SQLE ( Santa Cruz , sc-49754 , 1∶500 ) , anti-HMGCS1 ( Santa Cruz , Sc-32422 , 1∶500 ) , mouse anti-mCMV IE1 ( Chroma 101 , 1∶1000 ) , and rabbit anti-β-actin ( Cell Signalling , 4970 , 1∶2500 ) diluted in PBST ( 0 . 1% Tween20 ) . For secondary anti-goat IR-680 ( Invitrogen , A21088 , 1∶10 , 000 ) , IR-800 anti-mouse ( Thermo Fisher Scientific , 35571 , 1∶10 , 000 ) , and IR-800 anti-rabbit ( Cell Signalling , 5151 , 1∶10 , 000 ) , antibodies were diluted in Odyssey blocking buffer ( 0 . 1% Tween20 , 0 . 01% SDS ) . For probing , visualization , and quantification , the Odyssey protocol ( LI-COR ) was followed . The fluorescence was quantified by Odyssey system ( Li-COR ) . For details of anti-mouse SREBP-2 polyclonal antibody ( custom antibody raised against mature SREBP-2 form [67] ) and immunoblot procedures , see Text S1 . Intracellular cholesterol concentration was determined enzymatically using the Amplex-Red cholesterol Assay Kit ( Molecular Probes ) according to manufacturer recommendations . Briefly , cells were washed with 1 ml ice-cold PBS and then lysed in 200 µl cold Lipid buffer containing 0 . 5 M of potassium phosphate , pH 7 . 4 , 0 . 25 mM cholic acid , and 0 . 5% triton X-100 . Cell lysates were sonicated on ice with three 10-s pulses at high intensity . 20 µl were then used to determine protein concentration using a standard BCA assay to normalize the protein concentration . For cholesterol measurement , 20 µl of each sample were added to the 80 µl assay solution , which contained 300 µM Amplex Red reagent , 2 U per ml HRP and 2 U per ml cholesterol oxidase , 0 . 1 M of potassium phosphate , pH 7 . 4 , 0 . 05 mM cholic acid , and 0 . 1% triton X-100 . After preincubation for 30 min at 37°C under light exclusion conditions , fluorescence was measured using excitation at 530±2 . 5 nm and fluorescence detection at 590±2 . 5 nm with a Polarstar Optima Multifunciton Microplate Reader ( BMG Labtech , UK ) . The values were corrected from the background . The relative amount of free cholesterol to the mock-treated samples was calculated using the manufacturer's supplied standard curve . An Agilent high-performance liquid chromatography ( HPLC ) system coupled with an Applied Biosystem Triple Quadrupole/Ion Trap mass spectrometer ( 4000Qtrap ) was used for quantification of individual polar lipids ( Phospholipids and sphingolipids ) . Electrospray ionization-based multiple reaction monitoring ( MRM ) transitions were set up for the quantitative analysis of various polar lipids [68] . HPLC atmosphere chemical ionization/MS were carried out for analysis of sterols [69] . To measure the effect of multiple drugs and siRNA transfection on viral growth , a sensitive live cell infection assay was developed using the properties of the mCMV GFP tagged virus . 1 . 5×104 NIH/3T3 cells were infected for 1 h in black 96-well plates ( Costar , UK ) at an MOI of 0 . 2 in 25 µl of fresh DMEM phenol red-free media , 3% CS , and 100 U of penicillin/streptomycin per ml . After infection , the inoculums were carefully removed by pipetting and replaced by 150 µl of DMEM phenol red-free media with 10% FCS . Viral growth was measured by recording the GFP signal over time using an OPTIMA Polarstar plate reader ( excitation wave length of 485 nm and emission of 520 nm ) . As an optimization step we checked the correlation between GFP levels and MOIs . Results showed a good correlation between multiplicity of infection and growth kinetics ( Figure S4 ) . Comparing the GFP value and number of viral particles per ml using plaque assay checked levels of GFP signal corresponding to different levels of virus . Results ( unpublished data ) showed a strong correlation between differences in levels of GFP expression and differences in number of viral particles assessed by plaque assay: a drop of 20% of GFP signals corresponding to a log difference in the number of viral particles monitored by plaque assay . For transfection , siRNA ( SMARTpools-ON-TARGETplus modification ) from Thermo Fisher Inc . were purchased . The samples were supplied at a concentration of 5 mM and diluted and aliquoted in 2 µM amounts . To transfect at a final concentration of 20 nM per well , 1 µl of siRNA SMARTpool was used with 9 µl of Optimem ( Invitrogen , CA , USA ) solution while 0 . 4 µl of Dharmafect 1 ( Dharmacon , Perbio Science , Bonn , Germany ) was mixed with 9 . 6 µl Optimem . Following incubation for 5 min , the siRNA mix was added to the Dharmafect 1 ( 0 . 4% ) mix and incubated for a further 30 min , after which 1 . 5×104 NIH3T3 cells in 80 µl of DMEM 10% CS medium lacking antibiotics was added to the siRNA:Dharmafect 1 complexes . Growth medium was removed and cells were washed 1× in PBS before 100 µl of the siRNA: Dharmafect 1 liposomes were added . Transfection conditions were optimized by using siGLO RED from Thermo Fisher Scientific ( Dharmacon ) as an indicator of transfection efficiency and cell viability was assessed as described before . For every gene targeted , the knock-down efficiency was checked by QPCR after 48 h incubation . Each of the three genes targeted ( Hmgcs1 , Hgmcr , and Idi1 ) were knocked down by more than 70% , 48 h after transfection ( Figure S5 ) . Knock-down efficiency and cell viability were also checked at 5 d post-infection for the mCMV-GFP assay , and showed no alteration of the viability and a knock-down efficiency ≥50% ( unpublished data ) . BMDM were isolated and grown in the presence of Csf1 derived from L929 cells as described [63] except cells were cultivated in 15 cm diameter tissue culture plates for 7 d before treatment . Incorporation of 4-thiouridine ( Sigma ) into nascent RNA was undertaken as described [52] . In brief , at 360 min post-infection , 10 ml medium was aspirated from all plates , added to 80 µl 4-Thiouridine , mixed , and immediately returned to the culture dish . After 30 min , to end the RNA labeling period , terminate transcription , and lyse the cells , medium was aspirated from the labeled BMDM and replaced with 4 ml of RLT lysis buffer ( Qiagen ) . Total RNA was isolated using an RNeasy Midi kit ( Qiagen ) according to the manufacturer's instructions , quantitated using a Nanodrop ( Thermo Scientific ) , and integrity was confirmed using an Agilent Bioanalyser ( Agilent UK ) . Newly transcribed RNA ( ntRNA ) was then isolated as described in [52] and again quantitated using a Nanodrop . Processing of ntRNA samples ( 94 ng ) for hybridization to Affymetrix Mouse Gene 1 . 0 ST arrays was undertaken according to the manufacturer's instructions ( Affymetrix ) . Hybridisation , washing , staining , and scanning of the arrays were also undertaken following standard Affymetrix protocols . For the purposes of presentation , gene expression values for the specific genes of interest from control ( mock-infected ) BMDM were adjusted to a value of 1 . Values for expression in infected cells ( white ) were then expressed as a number relative to the control . All animal experiments had approval by the local animal ethics committee ( University of Edinburgh , Edinburgh , UK ) in accordance with recommendations of the Federation of European Animal Science Association and European legislation . Twelve mice ( C57/BL6 , Charles River , 12 wk of age ) were randomized into two groups of six animals each in two separate experiments . Simvastatin was prepared as described above . The dosages of statins used in the present investigation were chosen according to the literature [30] . At day 1 , mice were inoculated i . p . with 2×106 PFU per mouse . Animals were sacrificed 4 d post-infection . Spleen , liver , kidney , heart , and lung were harvested and sonicated as a 10% ( wt/vol ) tissue homogenate , and titers were determined by standard plaque assays , including centrifugal enhancement of infectivity on MEFp53−/− . The dashed line indicates the limit of detection ( 5×102 PFU/g ) . Horizontal bars indicate the median values . Normalization , filtering , statistical hypothesis testing for microarray data was carried out within the R Language and Environment for Statistical Computing ( www . r-project . org ) , using packages provided through the Bioconductor repository ( www . bioconductor . org ) . The majority of explorative analyses and visualizations were conducted with Partek ( Partek Incorporated , USA ) and GeneSpring GX ( Agilent ) . Statistical analyses on other data sources were performed in Microsoft Excel software . For real-time PCR and replication assay , all graphs represent the mean ± SD . An unpaired Student's t test was used for evaluation of statistical significance of real-time PCR . For in vivo experiment a Mann-Whitney U test was used . See Text S1 for statistical analysis of microarray experiments . Statistical significance: *p<0 . 05 , **p<0 . 01 , ***p<0 . 001 . | Currently , little is known about the crosstalk between the body's immune and metabolic systems that occurs after viral infection . This work uncovers a previously unappreciated physiological role for the cholesterol-metabolic pathway in protecting against infection that involves a molecular link with the protein interferon , which is made by immune cells and known to “interfere” with viral replication . We used a clinically relevant model based on mouse cytomegalovirus ( CMV ) infection of bone-marrow-derived cells . Upon infection these cells produce high levels of interferon as part of the innate-immune response , which we show in turn signals through the interferon receptor resulting in lowering enzyme levels on the cholesterol pathway . We observed this effect with a range of other viruses , and in each case it leads to a notable drop in the metabolites involved in the cholesterol pathway . We found that the control mechanism involves regulation by interferon of an essential transcription factor , named SREBP-2 , which coordinates the gene activity of the cholesterol pathway . This mechanism may explain clinical observations of reduced cholesterol levels in patients receiving interferon treatment . Our initial investigation into how lowered cholesterol might protect against viral infection reveals that the protection is not due to a requirement of the virus for cholesterol itself but instead involves a particular side-branch of the pathway that chemically links lipids to proteins . Drugs such as statins and small interfering RNAs that block this part of the pathway are also shown to protect against CMV infection of cells in culture and in mice . This provides the first example of targeting a host metabolic pathway in order to protect against an acute infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/viral",
"infections",
"virology",
"computational",
"biology/systems",
"biology",
"immunology/innate",
"immunity"
] | 2011 | Host Defense against Viral Infection Involves Interferon Mediated Down-Regulation of Sterol Biosynthesis |
The recommended strategy for control of schistosomiasis is preventive chemotherapy with praziquantel ( PZQ ) . Pre-school children ( PSC ) are excluded from population treatment programs . In high endemic areas , these children are also at risk , and require treatment with PZQ . The Government of Kenya initiated the National School-Based Deworming Programme ( NSBDP ) where PSC in Early Childhood Development Education ( ECDE ) Centers are only eligible for treatment with albendazole ( ABZ ) but not with PZQ . 400 PSC were enrolled , from 10 randomly selected ECDE Centers in Kwale County , Kenya where children were treated with crushed PZQ tablets mixed with orange juice , at a single dose of 40 mg/kg . Adverse events were assessed 24 hours post-treatment through questionnaires administered to the parents or guardians . Acceptability was determined by observing if the child spat and/ or vomited all or part of the PZQ dose immediately after treatment . Efficacy was assessed by examining urine samples for Schistosoma haematobium eggs in the 5 weeks post-treatment follow-up . Children testing negative for S . haematobium during the follow-up were considered cured . Egg reduction rate ( ERR ) was calculated as the decrement in the infection intensity ( group’s geometric mean egg counts per 10 ml of urine ) following treatment expressed as a proportion of the pre-treatment infection intensity . Before treatment , 80 out of the 400 children enrolled in the study tested positive for S . haematobium ( 20 . 0% ( 95% confidence interval ( CI ) 16 . 4–24 . 2% ) . Of these , 41 had infections of heavy intensity ( 51 . 3% ) while the rest ( 48 . 7% ) were of light intensity . Five weeks post-treatment , 10 children who had heavy intensity infection were diagnosed with S . haematobium ( prevalence: 2 . 5% ( 95% CI 1 . 5–4 . 9% ) . Infection intensities decreased significantly from 45 . 9 ( 95% CI: 31 . 0–68 . 0 ) eggs/ 10 ml urine to1 . 4 ( 95% CI: 1 . 1–1 . 7 ) eggs/ 10 ml urine during pre-and post-treatment respectively . The ERR was 96 . 9% . There were no severe adverse events during follow up 24 hours post treatment . Treatment tolerability among the 400 children was high as none of the children spat and/ or vomited as observed in this study . The study revealed that crushed PZQ is safe and effective in the treatment of urogenital schistosomiasis in this age group . It is therefore recommended that PZQ should be administered to the PSC in Kwale County .
Human schistosomiasis is a major neglected public health problem caused by trematodes of the genus Schistosoma . Over 200 million people are infected globally , with 85% of these cases living in Sub-Saharan Africa [1] . In Kenya , nearly 6 million people are infected and an additional 15 million are at high risk of infection particularly in endemic areas [2 , 3] . Schistosomiasis ( Bilharzia ) is classified as one of the neglected tropical diseases ( NTDs ) . These are a group of diseases found predominantly in tropical areas that are associated with poor sanitation and poverty and which have historically received insufficient attention towards their control . The majority of infections in sub-Saharan Africa are caused by S . mansoni and S . haematobium which reside in intestinal mesenteric veins and bladder respectively , leading to intestinal and urogenital schistosomiasis . In Kenya , S . haematobium occurs mainly in areas around the upper and lower Coast region and some parts of the Lake Victoria and Kano plains in Western Kenya [4] . In affected populations , children carry the heaviest burden of infection [5] , [6] . Symptoms of urogenital schistosomiasis include haematuria , dysurea , nutritional deficiencies , anemia , growth retardation , decreased physical performance and impaired memory and cognition [7–10 , 1] . Control of schistosome infections is through treatment of infected people with a single dose of the anti-helminth drug praziquantel ( PZQ ) which is safe , highly efficacious , cheap ( costing less than US$0 . 50/ dose ) and can reverse schistosome-related morbidity particularly in the early stages of disease progression [11] . Studies point to a growing body of evidence that in many endemic communities , schistosomiasis infection–contrary to previous beliefs–starts in early childhood . The presence of infection , points to the fact that infants and pre-school aged children are also at risk of infection like their older school-aged counterparts . The growing concern here is that infection in infants and pre-school children ( PSC ) may persist until the child starts school if left untreated . In preventive chemotherapy control programmes infants and PSC are not eligible for treatment until school-age [12–14] . Failure to reach a majority of the 2–6 year olds in Early Childhood Education ( ECDE ) Centers could result in higher prevalence of schistosomiasis and its negative health effects such as malnutrition and poor cognitive performance . In turn , these effects retard the child’s growth and development [15] . Treatment of children is also likely to be more successful in averting the development of subsequent , more serious disease sequelae because earlier stages of infection-induced pathology may be reversible if treated promptly [16] . World Health Organization ( WHO ) recommends that young children living in endemic areas be considered for treatment with PZQ during child health campaigns at the standard dose of 40mg/kg [12] . Current schistosome control programmes advocated by the World Health Assembly in 2001 through resolution 54 . 19 recommend regular de-worming of school age children at risk of infection with anti-helminthes [17] . However , these programs exclude pre-school age children due to the perception that these children are not sufficiently exposed to infective water to experience high infection rates [18] . One of the concerns associated with treatment of preschoolers for schistosomiasis is that they are believed to be at risk of choking on whole tablets . The other one is that there is limited formal data with respect to prescribing information by the pharmaceutical companies on toxicity , method of administration , adverse effects and pharmacokinetics in this age group [19] . This may result in clinical disease that is not managed and the lack of safety data on PZQ in this age group [11] . Previous studies have shown that , there have not been severe adverse reactions to PZQ treatment when given to young children due to the excellent safety and tolerability [17] . For administration of PZQ to children under 5 years , it is possible to break the tablet into small pieces or crush them in flavored syrup which would make the tablet palatable and acceptable [17] . The goal of this study was to assess the acceptability , adverse events and efficacy of treating pre-school children with praziquantel for S . haematobium infection in selected Early Childhood Development Education Centers of Kwale County , Kenya .
Permission to conduct the current study , including review and approval was obtained from the Scientific Steering and Ethical Review Committees , Kenya Medical Research Institute ( KEMRI ) SSC No . 2958 . The county education office , county health office , local leaders , teachers , children and parents/guardians were informed about the study in the area . Written informed consent was obtained from a parent or guardian , for every child in the study . In addition , assent was obtained from the children . This study was conducted in Kwale County , which is situated in the Coast Region of Kenya . It has 4 Constituencies: Msambweni , Lunga-Lunga , Matuga and Kinango . The total population stands at 649 , 931 , of which 36 , 197 are PSC [20] . Kwale County is mainly an inland county , but it has a coastline south of Mombasa . The area is hot and humid year round with annual mean temperature range of 22°C—34°C , average relative humidity range of 70% - 80% , and annual rainfall range of 900–1500 mm . Altitude ranges from 0 to 462 meters above sea level . The majority of the population 81 . 9% live in the rural areas with poor road and transport network . Poverty which stands at 71% and lack of sanitation in this area contributes a lot to the high prevalence of soil-transmitted helminthes especially in infants and pre-school children . A large proportion of the population in the study area has no access to safe water and adequate sanitation [21–22] The current study was conducted in Matuga and Lunga-Lunga constituencies . Under the Kenya National School Based Deworming Programme ( NSBDP ) , children in primary schools in the Coast region of Kenya were the first to receive treatment with albendazole and praziquantel . This was after results from a baseline survey in 2011 showed that the prevalence of soil transmitted helminthes and schistosomiasis was high [23] . This sub-study was embedded in a larger study Evaluating Different Drug Delivery Approaches for Treatment of Soil-transmitted Helminthiasis and Schistosomiasis Infections in the NSBDP among Children Attending ECDE Centers in Coast Province , Kenya . SC No . 2547 . In the above study , 28 ECDE Centers were targeted for treatment with PZQ . In the present study , 10 schools were randomly selected from these 28 ECDE Centers . All the children ≤ 6 years of age were enrolled in this study . The study sample was 400 PSC . This study was a longitudinal , pre and post-test design . Detection of Schistosoma infections was conducted before and after treatment with crushed PZQ mixed with orange juice . The acceptability and safety of PZQ was also assessed . The experimental design entailed laboratory examination of urine samples from the children , where efficacy of the crushed praziquantel mixed with orange juice was determined , by assessing the prevalence and intensity of the Schistosoma haematobium eggs pre and post treatment . The descriptive explanatory strategy assessed the acceptability of the crushed praziquantel mixed with orange juice . It also assessed any adverse events after treatment through researcher administered questionnaires to the parents/guardians of the ECDE children 24 hours after treatment . Acceptability was determined by observing if the child spat and/ or vomited all or part of the PZQ dose immediately after treatment . Any adverse events experienced by the children one hour post treatment were observed and recorded by the teachers of the ECDE children and community health extension workers ( CHEWs ) who took part in the treatment of the children . Eligibility for inclusion into this study included: 1 ) aged ≤ 6 years old at recruitment; 2 ) enrolled in ECDE centers that were targeted for treatment with Praziquantel; 3 ) production of a urine sample; 4 ) Parental/guardian consent to participate in the study . Participants who had existing medical conditions were excluded from the study . These criteria were based on the World Health Organization ( WHO ) Manual of Preventive Chemotherapy [17] Urine samples were collected from all the 400 enrolled children in clean labeled wide mouthed urine containers with lids , between 10 a . m . and 2 p . m . Visible hematuria was recorded upon urine collection . The labeled properly capped containers containing the urine samples were transported with a cool box to the KEMRI Center for Microbiology , Kwale Laboratory for examination . The urine was thoroughly mixed and a duplicate 10 ml aliquot of urine filtered through 15-mm polycarbonate filters ( Nuclear pore R; Costar Europe Ltd . , Badhoevedorp , the Netherlands ) . The filter paper was then placed on a labeled slide and a drop of Lugol’s solution added . The slides were then examined under a microscope within 6 hours and the mean counts of the two filters recorded and expressed as eggs per 10ml urine [17] . The intensity of infection was categorized according to the WHO classification as negative for no detectable eggs; light for 1–49 eggs/10 ml urine; or heavy for > 50 eggs/10 ml urine . 5 weeks post treatment , urine samples were collected from the children who had tested positive for ova of S . haematobium . This was to assess cure and egg reduction rates [17] . In the present study , a child was considered to have been cured if no S . haematobium eggs were detected microscopically in urine samples collected 5 weeks post-treatment . The egg reduction rate was calculated as the decrease in geometric mean intensities of S . haematobium eggs divided by pre-treatment geometric mean intensity multiplied by a factor of 100 . 400 PSC were enrolled , tested and treated for S . haematobium pretreatment . Praziquantel tablets ( Prazitel , Cosmos Ltd ) were used for treatment in this study . Each child was weighed using a calibrated weighing scale and a single dose of 40mg/kg PZQ administered . Before administration , the PZQ tablets after splitting were crushed with a mortar and pestle and the powder mixed with fruit juice to decrease the bitter taste . This was done during the health break after the children had eaten . Drug administration was supervised using the modified Direct Observation Therapy ( DOT ) . One hour post-treatment observations for any adverse events were made and recorded by the 10 ECDE teachers and CHEWs who took part in the deworming exercise . Parents or guardians of the treated children were also interviewed using structured questionnaires 24 hours post-treatment for episodes of treatment-related adverse events . The study clinician evaluated the following adverse events abdominal pain , dizziness , nausea , headache , vomiting , drowsiness , itching , as likely or unlikely associated with study drug . Other symptoms reported by parents or guardians were also recorded . Five weeks after praziquantel administration , urine samples from the children who tested positive for S . haematobium were collected again , using the same procedures . The efficacy of praziquantel was assessed five weeks post treatment using the same diagnostic criteria as baseline . This was determined by means of cure rate ( CR , percentage of children positive at the pretreatment cross-sectional survey who became egg-negative 5 weeks after treatment , as assessed by urine filtration for S . haematobium ) and egg reduction rate ( ERR , reduction in the group’s geometric mean S . haematobium egg count in 10 ml of urine comparing the before and after treatment situation ) [24] . In this study treatment acceptability was defined as the number of children spitting and/or vomiting all or part of the PZQ dose , immediately after treatment and it was assessed by DOT . This was the first time that these children were being treated in school with praziquantel . Data were double entered in Microsoft Excel spreadsheet . Statistical analyses were done with Statistical Package for Social Sciences ( SPSS version 17 ) . PSC who had at least one urine sample subjected to a filtration method for S . haematobium diagnosis before and after treatment were included in the final analysis . Continuous data ( e . g . , schistosome egg counts ) are presented as geometric mean . Infection intensities were stratified according to the cut-offs defined by the WHO [25] .
The present study enrolled a total of 400 children of preschool age ( ≤ 72 months ) . The mean age of the children was 4 . 8±1 . 1 years . Those aged three years or less constituting 11 . 3% , 4 years ( 26 . 3% ) , 5 years ( 27 . 5% ) and 6 years ( 35% ) of the study sample . Only one child was aged two years . Boys constituted 51 . 2% whereas girls were 48 . 8% of the enrolled children . Adverse events were assessed 24 hours post treatment . This was through researcher administered questionnaires to the parents and through observations made one hour post treatment by the ECDE teachers and CHEWs , who took part in the deworming exercise . 330 out of the 400 children recruited in the study were assessed for AEs . One experienced dizziness , one experienced a headache , four had abdominal pain/discomfort , two had nausea and two experienced itching . None of the children vomited . While six respondents took no action when their child experienced an adverse event , one gave food , two gave milk and the other one made the child to rest as shown in Table 3 . None of the 400 ( 100% ) PSC spat and/ or vomited during treatment . This was assessed by DOT , by ECDE teachers and CHEWs present during deworming .
Results of the current study showed that praziquantel achieved high cure rates of 86 . 2% against S . haematobium infections 5 weeks after treatment . This is in agreement with results of recent Cochrane systematic review which showed that treatment with the standard dose of praziquantel ( 40 mg/kg ) generally results in cure rates of 80% 1–3 months after treatment [24] . In this study , the results show that 10 children out of the 80 , who had tested positive for S . haematobium , had an infection of light intensity ( 1–49 eggs/10 ml urine ) after treatment . These were the children who had infections of heavy intensity of ( ≥50 eggs/10 ml urine ) , before treatment . The design of our study did not allow estimating the proportion of infections after treatment that were due to juvenile stages of the parasite , which are largely insensitive to praziquantel . This is in line with a study in Mali assessing urinary schistosomiasis in preschool aged children showing that , the presence of S . haematobium eggs five weeks post-treatment could be explained by factors such as high pretreatment worm load that could not be completely cleared by the treatment that remained in the treated children and started producing eggs , and the presence of high numbers of immature worms less sensitive to praziquantel that escaped drug action and matured to egg producing worms during subsequent follow-ups [26]; praziquantel is refractory against immature worms [24] . The effect of treatment in terms of egg reduction rates which was 96 . 9% was high and supported by evidence of Cochrane systematic review [5] . The results of this study revealed that the prevalence of S . haematobium in pre-school children from Kwale County was high ( 20% ) compared to that observed among school aged in the same county ( 24 . 5% ) [27] . The findings are consistent with emerging evidence that the burden of schistosomiasis is high in pre-school children . A similar study in Sudan investigating the safety , efficacy and acceptability of praziquantel in pre-school age children reported a prevalence of 31 . 1% [28] , which is higher than what was found in this study ( 20% ) . In Ghana , a study investigating the extent of schistosomiasis in pre-school children and infants found prevalence of 11 . 2% for S . haematobium , with the highest egg count detected in a 4-month old infant [28] , [29] . In a rural endemic area in Nigeria , prevalence of 58 . 1% was reported for S . haematobium in children aged 1–6 years [3] . Similar findings have emerged from Mali where prevalence of S . haematobium among pre-school children aged 1–4 years was found to be 51 . 2% [26] . In Uganda nearly 50% of children less than three years of age living along the northern shoreline of Lake Victoria had S . mansoni infections [13] . A recent study from the shoreline villages of Lakes Albert and Victoria in Uganda found even higher prevalence of S . mansoni ( 62 . 3% ) in pre-school children [13] . In Sudan , an earlier study found high prevalence of schistosome infection ( 40% ) among pre-school children in the Gezira Irrigation Scheme [30] . The common feature associated with infection in these children from the various settings would be likely due to the fact that the children and their caregivers ( parents or guardians ) share the common risk factor of proximity to large water bodies known to harbor infectious cerceria . Schistosomiasis in infants and pre-school age children is of concern for at least two reasons . First , this younger age-group plays a hitherto unrealized role in maintaining local disease transmission; even though these infected children may be excreting fewer eggs , it is their regular water contact that leads to contamination of water . Moreover , rinsing and washing children's soiled clothes in environmental water bodies also contributes towards more cryptic contamination and disease transmission [18] . Thus this age-group will play an increasingly important role in environmental transmission likely to frustrate the attempts made by preventive chemotherapy campaigns striving towards more general reductions in environmental transmission [31] . Second , such regular water contact is also likely to result in frequent ( re ) infection episodes , which lead to a progressive increase of individual worm burden . It is therefore likely that untreated infections acquired in early childhood contribute to worsening the longer-term clinical picture of disease in the individual . Lack of safe water supplies , inadequate sanitation , insufficient access to health care and prohibitive treatment costs all contribute to disease transmission and high morbidities , especially in infection with schistosomiasis . S . haematobium infection that is predominant in Coastal region of Kenya is found to cluster in a subset of school age children with suggestions of synergistic effects on anemia , cognitive performance and stunting[32] . Chronic anaemia during childhood is associated with impairment in physical growth , cognition , and school performance [33] , whereas severe anemia accounts for up to one half of the deaths in children younger than 5 years of age [34] . In Kenya , during the 2009 treatment , only primary school-age children , ( 6–14 years ) both enrolled and non-enrolled were covered by the National School Based Deworming Programme , leaving out the children in the age bracket of 2–6 years who attend the ECDE Centers or pre-school . This age bracket requires to be treated as they also carry a heavy worm burden and pose a risk of re-infecting the treated school-age children while interacting and playing at the community level . In Kenya , the population of children enrolled in ECDE Centers is 2 . 2 million [35] . A high percentage of infected children means that the environment becomes more heavily contaminated–which in turn increases the risk of infection for the whole community . By reducing the number of worms in children , everyone benefits [25] . Given the difficulties of younger children swallowing large PZQ tablets and an associated risk of choking , medications were administered in crushed tablet form and mixed with orange-juice as previously piloted [17] . Previous studies have shown that the fruit flavor helped to mask the bitter taste of PZQ [36] . In our study none of the children spat and/or vomited during treatment . Previous studies have shown that there have not been adverse reactions to PZQ treatment when given to young children due to the excellent safety and tolerability [37] . In this study the adverse events experienced by the children were not severe and included nausea , dizziness , vomiting , abdominal pain/discomfort , itching , and headache which were largely self-limited . This is in line with other studies involving pre-school children , where minor and transient side-effects 24 hours after treatment were reported in Uganda [30] , and in Mali [26] in accordance with established evidence that praziquantel is associated with minor and transient adverse events [26 , 36] In a study in Sudan there were no drug-related adverse events experienced after treatment with praziquantel [28] . In studies carried out in Uganda and Mali , the adverse events experienced were minor and transient 24 hours after treatment whereas in Sudan , no adverse events were reported .
In conclusion , the present study showed that crushed praziquantel administered to preschool children at a dose of 40 mg/kg is safe and effective in the treatment of urogenital schistosomiasis . The pre-school children experienced minor side effects which were temporal and most of them required resting under a shade until they subsided . The study also adds to the evidence base that , the prevalence of S . haematobium in preschool age children is high , and they should be regarded as high risk group in the area , and should be taken into consideration during treatment programs in Kwale County and other endemic regions . This will prevent long-term chronic ill-health or schistosomiasis-related complications later in life . | Control of schistosome infections is through treatment of infected people with a single dose of the anti-helminth drug praziquantel ( PZQ ) which is safe , highly efficacious , and can reverse schistosome-related morbidity particularly in the early stages of disease progression . However pre-school children are normally excluded due to the belief that these children are not sufficiently exposed to infective water to experience high infection rates . This could lead to clinical manifestation of the disease and the lack of safety data on praziquantel in this age group . Due to this we investigated the safety , efficacy and acceptability of praziquantel in Kwale County , Kenya . We examined urine samples from 400 preschool children . They were treated with crushed praziquantel ( 40mg/kg ) mixed with orange juice and the efficacy of the treatment was determined 5 weeks after treatment . Acceptability was determined by whether the child spat and/ or vomited the treatment through the direct observed treatment ( DOT ) . No child spat or vomited during treatment . Safety of the treatment was assessed by interviewing the parents of the treated children for adverse events ( e . g . , abdominal pain , dizziness , and headache ) . The treatment was well tolerated and most of the parasites were cleared by praziquantel . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusion"
] | [
"schistosoma",
"invertebrates",
"children",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"clinical",
"research",
"design",
"education",
"helminths",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"diseases",
"animals",
"urine",
"age",
"groups",
"research",
"design",
"neglected",
"tropical",
"diseases",
"africa",
"families",
"research",
"and",
"analysis",
"methods",
"schistosoma",
"haematobium",
"schools",
"adverse",
"events",
"people",
"and",
"places",
"helminth",
"infections",
"schistosomiasis",
"kenya",
"eukaryota",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"organisms"
] | 2018 | Safety, efficacy and acceptability of praziquantel in the treatment of Schistosoma haematobium in pre-school children of Kwale County, Kenya |
Infections with monkeypox , cowpox and weaponized variola virus remain a threat to the increasingly unvaccinated human population , but little is known about their mechanisms of virulence and immune evasion . We now demonstrate that B22 proteins , encoded by the largest genes of these viruses , render human T cells unresponsive to stimulation of the T cell receptor by MHC-dependent antigen presentation or by MHC-independent stimulation . In contrast , stimuli that bypass TCR-signaling are not inhibited . In a non-human primate model of monkeypox , virus lacking the B22R homologue ( MPXVΔ197 ) caused only mild disease with lower viremia and cutaneous pox lesions compared to wild type MPXV which caused high viremia , morbidity and mortality . Since MPXVΔ197-infected animals displayed accelerated T cell responses and less T cell dysregulation than MPXV US2003 , we conclude that B22 family proteins cause viral virulence by suppressing T cell control of viral dissemination .
Smallpox was among the deadliest infectious diseases in history and its eradication is a landmark in medicine . However , loss of orthopoxvirus ( OPXV ) -specific immunity facilitates the accidental introduction of zoonotic OPXV such as monkeypox virus ( MPXV ) and cowpox virus ( CPXV ) which cannot be eradicated due to animal reservoirs . This risk became evident during the first MPXV outbreak outside Africa , which occurred in the US in 2003 [1] . Although MPXV does not spread efficiently by human-to-human contact it shares several key features of pathogenesis with variola virus ( VARV ) the causative agent of smallpox . MPXV is endemic in African rain forests with strains circulating in Central versus West Africa falling into two genetically distinct clades [2] . The West African clade , including US2003 strains , is considered less virulent based on in vivo studies conducted in cynomolgus monkeys , prairie dogs , and ground squirrels [3] , [4] , [5] , [6] . Nevertheless , life-threatening disease was identified during the U . S . outbreak [1] , [7] . The DNA genomes of OPXV encode approximately 200 open reading frames ( ORFs ) with around 90 highly conserved genes encoded in the central regions of the genome whereas the terminally coded genes vary among different OPXV and are responsible for differences in host range , virulence , and immune evasion [8] . Conserved genes among OPXV are highly related to each other resulting in cross-protection , i . e . prior infection with any one of the OPXV generally protects against serious disease by other OPX , so that vaccinia virus ( VACV ) is broadly protective against all OPXV . Protection against OPXV is remarkably long lived . During the 2003 MPXV outbreak , the number of lesions in previously vaccinated individuals was significantly lower with some individuals being completely protected from MPXV-associated disease [9] . Antibody ( Ab ) titers to the vaccine remain remarkably stable over the life of vaccinated individuals [10] and vaccine-mediated protection of non-human primates ( NHP ) against lethal MPXV challenge is Ab-mediated [11] . Similarly , vaccinated mice succumb to lethal challenge with mousepox ectromelia virus ( ECTV ) in the absence of Ab , despite the presence of poxvirus-specific T cells [12] . In contrast , T cells promote survival of vaccinated mice challenged with lethal doses of VACV [13] , [14] . The limited role of T cells in protecting against virulent OPXV is surprising given that OPXV induce a strong T cell response recognizing multiple conserved epitopes [15] . Moreover , VACV is widely used as T cell-inducing vaccine vector [16] , [17] . The reduced ability of T cells to control OPXV might , in fact , be directly related to virulence since T cells do limit virulence of CPXV provided that two gene products interfering with MHC-I antigen presentation were deleted [18] . Thus , the inability of T cells in protecting against virulent OPXV might be due to T cell evasion mechanisms . In the case of CPXV , T cell evasion is mediated by two gene products that each interferes with different steps of the MHC-I antigen presentation pathway . CPXV203 binds to and retains MHC-I in the endoplasmic reticulum ( ER ) [19] . CPXV12 inhibits TAP-dependent peptide translocation across the ER membrane [18] , [20] . MPXV contains a CPXV203 orthologue , but does not seem to retain MHC-I [21] . Instead , MPXV inhibits T cell activation by either MHC-dependent or by MHC-independent stimulation [21] . Thus , MPXV encodes one or more gene products that render T cells non-responsive . Here , we identify the gene product responsible for this T cell inactivation as MPXV197 , the largest gene in the MPXV genome . This predicted transmembrane protein belongs to the B22 family of proteins found in several OPXV including CPXV , ECTV and VARV , but not in VACV . We demonstrate that MPXV197 and related proteins of CPXV and VARV inactivate T cells by a novel mechanism . We further show that deletion of MPXV197 severely attenuates MPXV and prevents lethal disease in rhesus macaques ( RM ) . Despite a substantial reduction of viral titers , RM infected with MPXV197-deleted virus had stronger and more rapid T cell responses consistent with B22 proteins contributing to OPXV virulence by suppressing T cell responses .
We previously demonstrated that MPXV ( Zaire strain ) inhibits CD4+ and CD8+ T-cell activation by both MHC-dependent and MHC-independent stimuli [21] . In contrast , T cell evasion by CPXV seemed to rely predominantly on inhibition of MHC-I-dependent antigen presentation by CPXV12 and CPXV203 [18] , [19] , [20] , [22] . Since MPXV Zaire encodes an orthologue of CPXV203 we wanted to determine whether it is required for T cell evasion . The West-African strain MPXV US2003 lacks most of the CPXV203 orthologue [2] . We compared poxvirus-specific T cell responses elicited by MPXV Zaire and US2003 by infecting human PBMC from recently VACV-vaccinated donors with MPXV at an MOI of 0 . 3 and analyzed T cell responses by intracellular cytokine staining ( ICCS ) for TNFα+ IFNγ+ cells . VACV was used as control since it does not inhibit T cell stimulation [21] . VACV vigorously stimulated virus-specific IFNγ+TNFα+ CD4+ and CD8+ T cells whereas ≤6% of either response occurred with MPXV Zaire or MPXV US2003 ( Fig . 1A ) . Total IFNγ+ T cell responses or total TNFα+ T cell responses were reduced to the same level as that observed with IFNγ+TNFα+ T cells ( data not shown ) . This lack of T cell stimulation was not due to lack of cross-reactivity or reduced rates of infection since VACV-specific T-cells recognize cells infected with UV-inactivated MPXV[21] and MPXV and VACV infected comparable numbers of cells in PBMC ( data not shown ) . Thus , the MPXV homologue of CPXV203 is not required for inhibition of poxvirus-specific T cells . To further examine whether the previously reported MHC-independent activation of T cells was also inhibited in the absence the CPXV203 orthologue we compared T cell activation by plate-bound αCD3 Ab in the presence of MPXV Zaire and US2003 . As shown in Fig . 1A ( right panel ) , MPXV US2003 retained the capability to inhibit MHC- independent T cell stimulation . T cell inhibitory genes are thus conserved in both clades of MPXV but absent or non-functional in VACV . Since in PBMC , OPXV infect CD14+ monocytes but rarely T cells we concluded that T cell interference by MPXV is due to a trans-inhibition [21] . However , to rule out with certainty that MPXV inhibits T cells in cis we separated MPXV-infection from antigen presentation by infecting human foreskin fibroblasts ( HFF ) with MPXV and co-incubating these cells with rhesus macaque ( RM ) -derived T cell lines specific for the MaMu-A*01-restricted SIV GAG181-189 epitope CM9 [23] . As APC we used autologous B cells immortalized by simian lymphocryptovirus ( BLCLs ) [23] . Thus , in this assay the infected cells ( HFF ) do not contribute to T cell stimulation which is provided by peptide-pulsed BLCLs . To prevent dissemination of the virus from HFF to T cells we took advantage of the fact that compound ST-246 inhibits egress of viral particles from infected cells [24] . Control experiments demonstrated that ST246 efficiently ( ∼90% ) prevented spread of VACV , CPXV , and MPXV to Jurkat T cells ( Fig . S1 ) . When ST-246-pretreated HFF were infected with MPXV , T cell stimulation by CM9 peptide-pulsed BLCLs was still inhibited to <10% of the uninfected cell control ( Fig . 1B ) confirming that MPXV inhibits T cell activation in trans . Since the T cell inhibitory factor is not secreted [21] this process most likely involves cell to cell contact . To identify the T cell evasion gene ( s ) we generated four deletion mutants each lacking about 10 kb in the termini of the MPXV US2003 genome ( Fig . 1C; Fig . S2A ) . Each of the mutants was examined for its ability to inhibit stimulation of T cells in PBMC from VACV-immune subjects ( Fig . 1D ) or peptide-stimulation of CM9-specific T cells from RM ( Fig . 1E ) . Mutants lacking ORFs 11–25 , 26–35 ( data not shown ) , or 184–193 did not activate poxvirus-specific T cells ( Fig . 1D ) and still inhibited peptide-stimulation of CM9-specific T cells ( Fig . 1E ) . In contrast , MPXV lacking ORFs 194–197 activated both CD4+ and CD8+ T cells in VACV-immune PBMC ( Fig . 1D ) and no longer inhibited peptide stimulation of CM9-specific T cells ( Fig . 1E ) . These data suggested that the MPXV194–197 genomic region encodes the T cell inhibitor . Since it seemed likely that T cell evasion would be mediated by a membrane bound extracellular protein , we deleted MPXV197 which is predicted to encode a large TM protein . As shown in Fig . 1D and E , MPXVΔ197 stimulated poxvirus-specific CD4+ and CD8+ T cells similar to VACV and peptide stimulation of CM9-specific T cells was no longer inhibited . Therefore , we conclude that MPXV197 is essential for T cell inhibition by MPXV . To determine whether ectopically expressed MPXV197 inhibits T cell stimulation , we inserted a codon-optimized version into plasmid and adenovirus expression vectors . MPXV197 is the largest ORF in the genome of MPXV encoding for 1880 amino-acids with a predicted molecular mass of 212 kDa , a predicted cleavable N-terminal signal peptide ( SP ) , multiple N-glycosylation sites , a C-terminal transmembrane ( TM ) domain , and potentially one or more internal TM domains ( Fig . 2A ) [2] . Transient expression of MPXV197 in CHO cells and immunoblotting with αFLAG-Ab revealed two predominant bands with apparent molecular mass of ∼150 kDa , and ∼140 kDa and several minor , smaller bands as well as a large protein >250 kDa ( Fig . 2B ) . To determine which of these proteins were located at the cell surface we performed surface biotinylation followed by streptavidin-precipitation and immunoblot with αFLAG antibody . The ∼150 kDa species was the predominant species in this assay ( Fig . 2C ) . Pulse-chase labeling revealed that the ∼150 kDa protein was a processing product derived from the large >250 kDa precursor protein . A minor ∼140 kDa fragment carrying the C-terminal Flag-tag was synthesized simultaneously with the large precursor protein suggesting that this fragment is derived from an internal start site ( Fig . 2D ) . Interestingly , Endoglycosidase H ( EndoH ) –treatment reduced the apparent molecular mass of the largest and the smaller fragment whereas the ∼150 kDa processing product was EndoH-resistant . This result suggests that the full-length protein is processed into a ∼150 kDa fragment that is transported beyond the ER to the cell surface consistent with the surface biotinylation result . The C-terminal location of the FLAG-tag identifies the ∼150 kDa fragment as a C-terminal fragment . We further determined the sub-cellular localization of MPXV197 by immunofluorescence analysis ( IFA ) using confocal laser scanning microscopy ( CLSM ) . Staining with αFLAG Ab of permeabilized or non-permeabilized CHO cells revealed that the C-terminus of MPXV197 locates to the extracellular face of the plasma membrane ( Fig . 2E ) . In contrast , N-terminally Flag-tagged MPXV reacted with αFLAG Ab only when cells were permeabilized ( Fig . 2E ) consistent with the full-length protein and potential N-terminal fragments remaining intracellular . The extracellular location of the C-terminus suggests that the C-terminal fragment most likely displays a multi-transmembrane topology with additional parts being exposed extracellularly ( Fig . 2A ) . The fate of the remaining N-terminal part of MPXV197 is currently unknown and additional work will be required to delineate the topology of this protein in more detail . To test whether MPXV197 inhibits T cell stimulation we co-incubated Ad-197 -transduced CHO cells with CM9-specific CD8+ T cells stimulated with peptide-pulsed BLCL . Thus , T cells were stimulated by exposure to cognate peptides presented by BLCLs whereas MPXV197 is provided in trans by expression in CHO cells . Since MPXV197 is under control of the tetracycline-regulated transactivator ( tTA ) CHO cells were co-transduced with Ad-tTA . CHO cells transduced with Ad-tTA alone did not inhibit T cell activation with cognate antigen ( Ad-control , Fig . 3A ) . In contrast , T cell responses were reduced to ∼0 . 01% of control upon MPXV197 expression ( Ad-197 , Fig . 3A ) . Thus , MPXV197 inhibits T cell stimulation in trans even when provided by unrelated cells of a different species . To measure the kinetics of the CM9-specific CD8+ T cell inactivation we co-incubated CM9-specific T cells with MPXV197-expressing cells for variable time periods prior to stimulation with peptide pulsed BLCLs . T cell stimulation was reduced following as little as 1h of exposure to MPXV197 , with maximal inhibition at 6 h of co-incubation ( Fig . 3B ) . Next we examined whether MPXV197 inhibits human T cell clones restricted by either classical ( HLA-B ) or non-classical ( HLA-E ) MHC-I using M . tuberculosis-specific CD8+ T cell clone D466 D6 recognizing peptide CFP2-12 presented by HLA-B [25] and D160 1-23 which is stimulated by pronase digested Mtb cell wall in the context of HLA-E [26] . BEAS-2B epithelial cells were infected with either Ad-197 alone or together with Ad-tTA followed by incubation with Mtb-specific CD8+ T cell clones . As shown in Fig . 3C , stimulation of both clones was inhibited by MPXV197 . Since in this assay , MPXV197 is expressed in the same cells that present antigen we additionally examined antigen- and MHC-independent T cell stimulation of these T cell clones by phytohaemagglutinin ( PHA ) , a lectin that activates the TCR non-specifically by carbohydrate cross-linking . PHA stimulation of both D466 D6 and D160 1–23 was inhibited by MPXV197 . Taken together , these data demonstrate that MPXV197 recapitulates the T cell unresponsiveness mediated by MPXV for both human and non-human primate T cells regardless of the TCR stimulus . Lack of cellular amine-reactive fluorescent staining ( LIVE/DEAD Fixable Dead Cell Stain ) indicates that T cell membranes remain intact in the presence of MPXV197 ( Fig . 3D , right panel ) . To further determine whether T cells would be activated upon by-passing TCR stimulation we stimulated CM9-specific T cells with phorbol 12-myristate 13-acetate ( PMA ) which activates protein kinase C ( PKC ) and the Ca2+ ionophore ionomycin ( Iono ) . Unlike peptide stimulation , MPXV197-expressing CHO cells did not inhibit T cell stimulation by PMA/Iono ( Fig . 3D , left panel ) . Thus , T cells remain viable after exposure to MPXV197 suggesting that MPXV197 counteracts TCR-dependent signal transduction upstream of PKC . Moreover , exposure of CM9-specific CD8+ T cells to MPXV197-expressing CHO cells did not impair their ability to bind a MaMu-A*01/CM9 tetramer suggesting that MPXV 197 does not interfere with MHC-I peptide loading ( Fig . 3E ) . We conclude that MPXV197 interference occurs after TCR engagement with peptide/MHC complexes most likely inhibiting TCR proximal signal transduction . MPXV197 belongs to the B22- protein family found in several OPXVs ( Fig . 4A ) including CPXV ( CPXV219 , 84% amino-acid identity ) , which causes zoonotic infections in humans , and VARV ( B22 , 86% amino-acid identity ) , the causative agent of smallpox . To examine whether VARV B22 also inhibits T cell stimulation we inserted codon-optimized B22 encoding for 1897aa with a predicted molecular mass of ∼214 kDa into expression vectors . Similar to MPXV197 , immunoblots and surface biotinylation of VARV B22 revealed surface expressed ∼150 kDa fragment with the B22 fragment being slightly larger than the corresponding MPXV197 fragment ( Figs . 4B , C ) . Also similar to MPXV197 was the observation that the full-length precursor protein was barely detectable at steady state consistent with the 150 kDa protein being the final product . The smaller protein bands were less abundant than those seen in MPXV197-expressing cells . It is possible that these minor bands in both MPXV197 and VARV B22 are by-products of high level ectopic expression . Similar to MPXV197 , we observed that the C-terminus of VARV B22 is exposed at the cell surface ( Fig . 4D ) . T cell inhibition by VARV B22 was examined using both human Mtb-specific CD8+ T cell clones and rhesus CM9-specific CD8+ T cell lines . As shown in Fig . 4E and F , VARV B22 inhibited T cell stimulation of both human and RM T cells as efficiently as MPXV197 . These data strongly suggest that VARV inactivates T cells in a manner that is similar to MPXV . To determine whether CPXV219 would similarly inhibit T cells we used a recombinant VACV expressing CPXV219 to infect BEAS-2B cells and examine stimulation of Mtb-specific T cells or to infect HFF and monitor stimulation of SIV-specific T cells by CM9 peptide loaded BCBLs . Whereas VACV did not impact stimulation of human or RM T cells , VACV-219 inhibited T cell stimulation in both instances ( Fig . 5 A , B ) . Thus it seems that the B22 proteins represent a family of T cell inhibitors . The finding that CPXV219 inhibits T cells was unexpected since we previously reported that poxvirus-specific T cells were stimulated once MHC-I-dependent antigen presentation by CPXV was restored due to deletion of CPXV12 and CPXV203 [20] . However , genome analysis of our deletion virus CPXVΔ12Δ203 revealed that , upon passaging , this mutant had acquired additional deletions downstream of CPXV203 due to a recombination event resulting in ORF204-221 being replaced by a duplication of ORF10-11 ( data not shown ) . Therefore , this deletion virus ( now designated CPXVΔ12Δ203-221 ) lacks not only CPXV12 and CPXV203 , but also CPXV219 . However , an independently generated CPXVΔ12Δ203 mutant was also reported to stimulate poxvirus-specific T cells [18] , although this analysis was limited to murine T cells . To determine the impact of CPXV219 on human and mouse T cells we analyzed CPXV mutants lacking CPXV219 alone or together with CPXV12 and CPXV203 and mutants lacking CPXV12 and CPXV203 . Stimulation of poxvirus-specific human T cells was analyzed by infecting PBMC from VACV-immune subjects with CPXV and monitoring T cell activation by ICCS whereas stimulation of murine T cells was monitored by adding splenocytes from VACV-immunized mice to CPXV-infected A20 cells ( Fig . 5C , D ) . CPXV did not stimulate poxvirus-specific human CD8+ and CD4+ T cells whereas mutant A694 lacking the genomic region CPXV204-221 stimulated human CD4+ T cells but not CD8+ T cells ( Fig . 5C ) . Since A694 contains CPXV12 and CPXV203 these data suggest that human CD8+ T cells are not stimulated due to MHC-I evasion whereas human CD4+ T cells were stimulated due to the absence of CPXV219 . Indeed , CPXVΔ12Δ203-221 lacking CPXV219 as well as CPXV12 and CPXV203 restored stimulation of both human CD4+ and CD8+ T cells . In contrast , CPXVΔ12Δ203 did not restore stimulation of poxvirus-specific human CD8+T cells ( Fig . 5C ) , despite the restoration of MHC-I presentation [18] . These results are consistent with CPXV219 inhibiting both human CD4+ and CD8+ T cells in a manner similar to MPXV197 . However , when stimulation of murine poxvirus-specific T cells was examined with the same series of mutants , CD8+ T cells were stimulated in the absence of CPXV12 and CPXV203 even when CPXV219 was present ( Fig . 5D ) as reported [18] . Interestingly however , CPXVΔ12Δ203 showed reduced activation of CD4+ T cells compared to VACV or CPXVΔ12Δ203-221 suggesting that CPXV219 does not efficiently inactivate murine CD8+ T cells but might impact murine CD4+ T cells . Together with the finding that MPXV stimulated murine CD8+ T cells ( data not shown ) these results indicate that B22 proteins inhibit human and monkey T cells , but are less active against murine T cells . Using a rabbit anti-serum raised against purified GST-tagged CPXV219 we examined its expression in CPXV-infected human 143 cells and CHO cells as well as in HEK 293 cells infected with VACV-219 ( Figs . 5E , F ) . CPXV219 was expressed with early kinetics and detectable as early as 3h p . i . ( data not shown ) . Metabolic pulse/chase labeling and immunoprecipitation at 3 h p . i . further demonstrated that a high molecular mass product ( >220 kDa ) was processed into a ∼150 kDa fragment ( Fig . 5E ) . Consistent with the ∼150 kDa fragment being the final product , a similarly sized protein was the predominant fragment in immunoblots of CPXV-infected CHO cells whereas this was absent from CPXVΔ219-infected cell lysates ( Fig . 5F ) . Similarly , a ∼150 kDa fragment was the predominant protein found in VAC-219 infected cells in the absence of the T7 polymerase . However , upon co-infection with T7-polymerase expressing VACV , the >250 kDa precursor was highly expressed whereas the ∼150 kDa fragment was only slightly increased consistent with the majority of the protein remaining in the ER-resident precursor state upon overexpression . Taken together with the data shown above for MPXV197 and VARV B22 , these data suggest that in both virally infected and ectopically expressing cells the full-length precursor protein is processed into a ∼150 kDa fragment . Since the anti-CPXV219 antiserum was raised against the whole protein , it is not known which part of the protein is recognized . However , the ∼150 kDa fragment of both MPXV197 and VARV B22 was detected by a C-terminal FLAG-tag suggesting that the CPXV219 ∼150 kDa fragment is likewise C-terminal . Thus , we conclude that a ∼150 kDa C-terminal fragment is the ultimate product of MPXV197 , VARV B22 and CPXV219 and that this fragment is transported to the cell surface where it acts as T cell inactivator . Since B22 proteins are more active against primate than rodent T cells we used a recently described intrabronchial ( i . b . ) inoculation model in RM [27] to determine the role of MPXV197 in viral dissemination , pathogenesis and induction of T cell responses . To rule out that MPXVΔ197 contained additional mutations compared to parental strain MPXV-US2003 we sequenced the genomes for both viruses by next generation ( NextGen ) sequencing . Within a margin of error ( <3% ) both WT and MPXVΔ197 matched the predicted sequence exactly ( Fig . S3 , Table S2 ) . Since this analysis cannot distinguish between sequencing errors , misalignments ( particularly in the repeat region ) and actual mutations , it is likely that the actual percentage of correct genome sequences is substantially higher . Therefore , we conclude that the vast majority of genomes present in our WT control and MPXV197-deleted virus contain the expected genome sequence . We infected 8 RM with MPXV-US2003 or MPXVΔ197 using i . b . inoculation of 2×105 PFU ( Fig . 6A ) , a dose at which MPXV-Zaire was non-lethal [27] . The clinicopathologic course of infection was followed by physical examination , biotelemetry to record body temperature and activity , O2 tissue saturation , and development of cutaneous lesions . Blood and bronchoalveolar lavage ( BAL ) fluid samples were collected at defined days post infection ( dpi ) to determine the kinetics of virus replication and of the adaptive immune response . As shown in Fig . 6 and table S3 , RM infected with MPXVΔ197 experienced a significantly shorter duration of fever ( 5 days compared to 20 days ) ( Fig . 6B ) , fewer skin lesions ( Fig . 6E ) , and dramatically reduced morbidity and mortality . In fact , two of the MPXV-US2003-infected RM had to be euthanized due to deteriorating health whereas all four of the MPXVΔ197-infected RM spontaneously controlled the infection prior to termination of the experiment at days 41 and 42 . Viral titers measured in the lungs were initially similar , reflecting the similar size of the inoculum , but lung titers of MPXVΔ197 fell significantly more rapidly compared to WT ( Fig . 6C ) . An even more striking contrast was observed for viral titers in the blood where all RM infected with MPXV-US2003 showed significantly higher levels of viremia compared to MPXVΔ197 which was barely detectable ( Fig . 6D ) . Interestingly , while uncontrolled viremia in both lungs and blood correlated with rapid deterioration of health in one animal ( WT-4 ) , the other animal that needed to be euthanized prematurely ( WT-3 ) had a lower viremia in the blood but a higher number of lesions at days 14 and 21 compared to the remaining WT-infected RM ( Fig . 6E ) . In contrast , low titers in the blood correlated with a generally mild disease and less than 30 lesions in MPXVΔ197-infected RM ( Fig . 6E , Table S3 ) . Decreased viral titers of MPXVΔ197 were also reflected in a decrease of antibody titers which tended to be lower than that of MPXV-US2003 although this was not statistically significant ( Fig . 6F ) . In stark contrast to the reduced virologic and disease parameters , poxvirus-specific T cell responses were detected earlier and were significantly higher at some of the earliest time points in RM infected with MPXVΔ197 compared to MPXV-US2003 ( Fig . 7A ) . ( Note that T cell responses were measured using VACV to avoid the T cell inhibitory effect of MPXV197 ) . At day 14 , all four MPXVΔ197-infected RM had a significantly higher frequency of poxvirus-specific CD8+ T cells in their blood compared to the 3 remaining WT-infected RM ( Fig . 7A ) . Similarly , in 3 of 4 MPXVΔ197-infected RM the CD4+ T cell response was above background at days 7 and 14 whereas 0/4 or 2/3 WT-infected RM had detectable CD4+ and CD8+ T cells at these days . At day 21 , the frequency of CD4+ T cells in all MPXVΔ197-infected RM was significantly higher than in WT-infected RM . The inverse correlation between viral titers and T cell responses in the blood is consistent with MPXV197 contributing to viral dissemination during the early phase of infection by delaying the onset of the cellular immune response . To examine whether the T cell inactivation mediated by MPXV197 would result in a systemic suppression of T cell responses during viral infection in vivo , we stimulated T cells in PBMC with αCD3 Ab . The data is limited to three WT and two MPXVΔ197-infected RM since two animals were missing samples and T cells from Δ197-3 was unresponsive to αCD3 stimulation potentially due to CD3 polymorphism . Although the overall frequency of T cells in the blood did not change during infection ( Fig . 7B ) , there was a dramatic reduction in αCD3 responses of both CD4+ and CD8+ T cells from WT-infected RM at 7–21 dpi ( Fig . 7C ) . This was particularly evident at day 14 which correlated with peak viremia in the blood of WT-1 and WT-2-infected animals ( Fig . 6D ) . In contrast , this decrease was less pronounced for αCD3-stimulation of T cells in both MPXVΔ197-infected RM . Although not statistically significant due to the low sample size , these observations are consistent with MPXV197 contributing to a systemic suppression of T cell responses during peak viremia .
We report here a novel mechanism by which viral proteins inhibit T cell responses and describe the impact of this immunomodulation on viral virulence and immunity . The activation of T cells via TCR engagement with peptide/MHC complexes is the principal mechanism by which CD8+ T cells recognize and eliminate virus-infected cells and by which CD4+ T cells recognize APC . To limit T cell control , viruses can thus either interfere with antigen presentation or they can interfere with the ability of T cells to respond to antigen . Many instances of the former mechanism have been described , particularly for large DNA viruses that are limited in their ability to escape from T cell control by mutating immunodominant epitopes [28] . However , MHC-inhibitors are unable to prevent activation of T cells by non-infected APC processing exogenous antigens via MHC-II presentation or by MHC-I cross-presentation . In contrast , B22 proteins would allow poxviruses to interfere with TCR-dependent T cell activation regardless of the stimulatory pathway . While indirect inhibition of T cell stimulation , e . g . by interference with cytokine networks [29] , [30] , [31] , [32] has been described for poxviruses , very few instances of viruses directly inhibiting T cells have been reported to date . Among poxviruses , a secreted CD30 homologue of ECTV was shown to inhibit CD4+ T cells activation in mixed lymphocyte reactions [33] . However , the CD30 homologue encoded by the CPXV strain used in our studies did not seem to affect T cell stimulation since both human and mouse T cells were activated upon elimination of the MHC-I inhibitory genes CPXV12 and CPXV203 and the T cell inactivator CPXV219 . It was also reported that VACV inhibits TCR-dependent responses of γδ T cells [34] , but this inhibitory mechanism clearly does not apply to αβ T cells since we used VACV as our control in all T cell experiments . To our knowledge , the most closely related mechanism described so far for a viral protein interfering with T cell stimulation was reported for UL11 of human cytomegalovirus ( HCMV ) which binds to CD45 on the surface of T cells and inhibits TCR-mediated signaling [35] . However , UL11 is unable to prevent the stimulation of HCMV-specific T cells by fibroblasts infected with HCMV lacking viral MHC-I inhibitors [36] . In contrast , deletion of CPXV MHC-I inhibitors did not restore human T cell stimulation when CPXV219 was present ( Fig . 5C ) . Moreover , MPXV seems to rely solely on MPXV197 for T cell inhibition . Thus , the strong inhibitory effect of B22 proteins stands out since it seems to require only a brief interaction with T cells , even non-cognate T cells , to shut down T cell stimulation . Since effector T cells are exquisitely sensitive to activating signals , one peptide/MHC complex reportedly can activate cytokine secretion in T cells [37] , the T cell inhibition by B22 family proteins is truly remarkable . This silencing of T cells is likely transient since T cells respond to stimuli that by-pass TCR signaling such as PMA/Ionomycin in the presence of MPXV197 ( Fig . 3D ) . The B22 inhibitory impact could thus be best described as transient inactivation , or transient “functional paralysis” , a term previously suggested for HCMV UL11 [35] . The molecular mechanism by which B22 proteins inactivate T cells is currently unknown . A multitude of proteins expressed on T cells can exert negative signals , in many cases overriding the ability of the TCR to signal . Engagement of these inhibitory receptors results in peripheral T cell tolerance that can be either permanent , as in T cell exhaustion , or temporary . Therefore , these proteins are often referred to as immune checkpoint proteins or co-inhibitory proteins [38] . Conceivably , B22 proteins could exert their effect by engaging one or several of these co-inhibitory proteins . Since B22 proteins seem to be unable to block murine CD8+ T cells it seems that the target protein on T cells is either primate-specific or that it is differentially expressed on T cells in mice . The inability of these proteins to inhibit murine T cells might be the reason why this effect has not been observed previously when studying CPXV in mice . Since ECTV is a mouse pathogen it will be interesting to examine whether the ECTV B22 homologue is capable of inhibiting mouse T cells . If not , the question arises whether B22 proteins perform a different function in rodents , or whether the lack of T cell inhibition is only observed in the genus mus , but not in other rodent species . Since both CPXV and MPXV infect rodents in the wild , it would be interesting to explore the species-specificity of this protein family in more detail . Additionally , B22 proteins might perform different functions in different species . For instance it is conceivable that in addition to the membrane bound ∼150 kDa C-terminal fragment , a secreted amino-terminal fragment is generated that has functions other than T cell inactivation . In contrast to MPXV and CPXV which display a rather wide host range , VARV is highly restricted to humans which enabled the eradication of this virus . The finding that VARV B22 inhibits human T cells , together with the finding that the corresponding protein in MPXV contributes dramatically to virulence , suggests that the presence of B22 might have contributed to the devastating pathology of smallpox infection . It is estimated that smallpox caused as many as 300–500 Million deaths in 20th century alone [39] rendering this virus one of the most , if not the most , virulent infectious disease ever to have affected the human population . Our data suggest that T cell evasion by B22 potentially contributed to this deadly outcome . MPXV197 is the first MPXV protein that was shown to have such a dramatic impact on virulence . Previous observations demonstrated that deletion of a viral complement modulator from MPXV-Zaire slightly increased the virulence in RM [27] . Of note , the mutant virus used in this previous experiment contained the same GFP/GPT expression cassette as the one used to replace MPXV197 thus eliminating the possibility that the attenuation of MPXVΔ197 was due to the presence of these heterologous gene product which were absent from the control virus . Interestingly , the complement binding protein is absent from West-African strains of MPXV including MPXV-US2003 used in our experiments as WT virus . The West-African clade is assumed to be less virulent than the central African clade in the human population [2] and the US2003 strain has been shown to be less virulent in small animal models of MPXV [3] , [4] , [5] as well as in a small study in cynomolgus macaques [6] . In our experiment , however , the US2003 strain was highly virulent since two of the four monkeys had to be euthanized when given a dose of 2×105 PFU . This mortality is similar , if not higher , than the mortality of RM infected with MPXV-Zaire at the same dose [[27] and unpublished observations] . Compared to the high morbidity and mortality observed with MPXV-US2003 , infection with MPXVΔ197 resulted in a comparatively benign disease that was controlled by the host . Viral load at the primary site of infection in the lungs was initially similar to WT , indicating that initial viral replication does not depend on MPXV197 consistent with MPXV197- deletion having no effect on viral replication in vitro ( Fig . S2B ) . In contrast , MPXV197 is required for efficient dissemination since both blood titers and skin lesions were reduced by orders of magnitude compared to WT . This effect correlates both with the timing and role of T cell responses in limiting poxviral disease . In WT-infected RM , T cell responses developed in 3 weeks and peaked around 4 weeks pi . In those animals that survived the infection with WT-virus , the development of measurable T cell responses coincided with control of viremia and reduction in fever that occurred between 2 and 3 weeks pi . By comparison , T cell responses to MPXVΔ197 were already detectable at 2 weeks pi and peaked around 3 weeks . In all MPXVΔ197-infected RM had returned to normal and viremia was no longer detectable by 10 dpi . The T cell response to MPXVΔ197 is thus similar to that observed for VACV which is clearly detectable at 2 weeks pi [16] . In contrast , the delayed T cell response to WT MPXV is likely mediated by MPXV197 which presumably contributes to the significant virulence of MPXV in the human host . The reduced spreading from lungs to blood is thus most likely an indirect effect of T cells more efficiently controlling the dissemination of MPXVΔ197 compared to MPXV US2003 . Previously it was shown that deletion of the MHC-I downregulating molecules CPXV12 and CPXV203 decreased CPXV mortality in mice [18] and deletion of MYX153 decreased virulence of myxoma virus in rabbits [40] . Taken together with our observations , these data suggest that the ability to limit T cell responses increases poxviral virulence in a number of poxvirus species . Our data also suggest that MPXV197 contributes to a temporary systemic immune suppression during acute infection . Systemic T cell suppression in vivo by an OPXV protein has not been previously observed . In contrast , generalized immune suppression mechanisms that involve peripheral T cell dysfunction are well known for chronic infection with viruses such as HIV , HBV , HCV , or chronic strains of LCMV [41] . However , T cell dysfunction in these cases is not the result of T cell shut-off by dedicated viral proteins but rather involves viral interference with immune regulatory networks , such as T cell exhaustion due to persistent antigen exposure or depletion of dendritic cells by viral infection [41] . Thus , systemic immune suppression by dedicated viral proteins inhibiting TCR signal transduction could represent a novel mechanism of viral T cell dysregulation . Such systemic T cell inactivation could also impact the priming of T cells since it would broadly inactivate the ability of T cells to respond to TCR-stimulation . T cell priming could either be delayed or the specificity of primed T cells could be skewed towards epitopes that are less affected by T cell inactivation , e . g . cross-presented epitopes versus epitopes presented by infected cells . The efficiency with which B22 proteins shut down T cells suggests that it might be possible to use such proteins to therapeutically reduce unwanted T cell-mediated inflammation . Using poxviral immunomodulators to control unwanted immune responses has already been demonstrated in a recent phase II clinical trial in which an anti-inflammatory poxviral protein of the Serpin-family significantly reduced myocardial damage biomarkers in patients receiving percutaneous coronary interventions such as coronary angioplasty or stent implantation [42] . This work establishes a proof-of-principle for the therapeutic use of viral immune modulators . Alternatively , B22 proteins might reveal novel inhibitory proteins on T cells or novel insights into T cell regulation that can be targeted to counter unwanted T cell responses . The precise mechanism of T cell inactivation by B22 proteins is therefore expected to provide novel insights into mechanisms of peripheral tolerance .
Human foreskin fibroblasts ( HFF ) , BEAS-2B human bronchial epithelium cells , human 143 cells , Chinese hamster ovary ( CHO ) cells , and human embryonic kidney ( HEK ) 293 cells were maintained in Dulbecco's modified Eagle's medium ( DMEM , Mediatech , Manassas , VA ) supplemented with 10% fetal bovine serum ( FBS , Hyclone Laboratories , Inc , Logan , UT ) . Rhesus macaque ( RM ) B-lymphoblastoid cell line ( BLCL ) was grown in 10% FBS-RPMI 1640 medium ( Hyclone Laboratories , Inc ) . Mtb-specific T cell clones and monkey CM9-peptide specific T cell lines and were maintained as previously described [23] , [25] , [26] . BSC40 , African Green Monkey kidney cells were grown in minimum essential medium ( MEM , Mediatech ) . Jurkat T cells clone JJK were grown in 10% FBS-RPMI 1640 medium ( Hyclone Laboratories , Inc ) . Vaccinia virus ( VACV ) Western Reserve strain , monkeypox virus ( MPXV ) strains Zaire and US2003 , Cowpox virus ( CPXV ) Brighton Red strain were propagated in BSC40 cells maintained in 5% FBS MEM . The virus preparations were purified using standard protocol [43] with minor modifications . Briefly , the infected cells were harvested , resuspended in 10 mM Tris-HCl ( pH 8 . 0 ) , and lysed by three cycles of freezing-thawing followed by two cycles of sonication . Precleared cell lysate was layered onto 36% sucrose cushion and centrifuged at 40 , 000×g for 80 min . Pelleted virus particles were resuspended in 1 mM Tris-HCl ( pH 8 . 0 ) and titered . For complete genome sequencing and in-vivo studies , the virus was additionally purified by centrifugation ( 22 , 500×g , 40 min ) through a 25% to 40% continuous sucrose gradient . VACV-immune subjects provided informed written consent before signing research authorization forms that complied with the US Health Insurance Portability and Accountability Act ( HIPAA ) in addition to a medical history questionnaire . These studies were approved by the Institutional Review Board of OHSU . All animal studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals ( 8th edition , The National Academies Press ) and the Animal Welfare Act ( the National Institutes of Health Office of Laboratory Animal Welfare assurance number A3304-01 ) . All animal procedures were performed according to protocols #0865 and #0731 approved by the Institutional Animal Care and Use Committee of the Oregon Health and Science University . Appropriate sedatives , anesthetics and analgesics were used during handling , and clinical and surgical procedures to ensure minimal pain , suffering and distress to animals . Female BALB/c mice at 5 months of age were purchased from The Jackson Laboratory . Mice were immunized intraperitoneally ( i . p . ) with 2×106 PFU/mouse of VACV WR . On day 8 post inoculation , spleens were collected and used for studies of the T cell responses to CPXV BR wild type and mutants . Eight adult female RM animals were utilized for in-vivo studies of the T cell responses to MPXV US2003 wild type ( WT ) and MPXVΔ197 mutant . Cohort 1 ( WT ) included animals 29437 ( WT-1; 7 year-old ) , 29785 ( WT-2; 10-year-old ) , 21111 ( WT-3; 13-year-old ) , and 28689 ( WT-4; 13-year-old ) . Cohort 2 ( Δ197 ) included animals 29792 ( Δ197-1; 8-year-old ) , 29398 ( Δ197-2; 11-year-old ) , 29424 ( Δ197-3; 13-year-old ) , 28664 ( Δ197-4; 10-year-old ) . The animals were infected intrabronchially with 5×105 PFU/animal of WT and the mutant viruses delivered in 1 ml of phosphate-buffered saline ( PBS ) . Blood and bronchoalveolar lavage ( BAL ) samples were collected on the day of infection ( day 0 ) and later on indicated days p . i ( Fig . 7A ) . Peripheral blood mononuclear cells ( PBMC ) were isolated from blood by centrifugation over Lymphocyte Separation Media . Body temperature and physical activity were monitored via telemetry implants ( Mini Mitter , Bend , OR ) . Codon-optimized sequences of the C-terminal 3×FLAG ( DYKDHDGDYKDHDIDYKDDDDK ) fusions of MPXV197 and VARV B22 proteins were synthesized at GenScript ( Piscataway , NJ ) . MPXV197 N-terminal Flag fusion was constructed by removing 3×FLAG sequence from the C-terminus of the protein and inserting it downstream of the predicted signal sequence after the amino acid E21 . All coding sequences were cloned into pCDNA3 . 1 vector ( Life Technologies ) . Additionally , MPXV 197-CFlag and VACV B22-CFlag coding sequences were sub-cloned in pAdtet7 shuttle vector [44] under a tetracycline ( tet ) regulated promoter . The resulting plasmids were used for construction of the recombinant adenoviruses viruses . To achieve the protein expression these viruses were co-infected with Ad-tTA virus expressing tet-transactivator ( tTA ) protein [45] . In vitro synthesis of VARV B22R ORF and all in vitro experiments using B22R-expressing constructs were approved by the World Health Organization ( WHO ) . BSC40 cells were plated into 6-well plates at 30% confluency . The next day , the cells were infected with 250 µl of a serial 10-fold dilution of the virus preparation or the infected cell lysate . At 1 h p . i . , the cells were overlaid with 0 . 5% agarose ( Life Technologies , Grand Island , NY ) -EMEM ( Quality Biological , Gaithersburg , MD ) and incubated for 5 days at 37°C . The cells were fixed with 75% methanol-25% Acetic Acid for 20 min and stained with 0 . 1% crystal violet -30% ethanol . Genomic DNA of the wild type MPXV and Δ197 mutant was isolated using DNeasy kit from the virus preparations purified through a 25% to 40% continuous sucrose gradient . DNA libraries were generated by the OHSU Massively Parallel Sequencing Shared Resource ( MPSSR ) core using the TruSeq DNA Sample Preparation kit ( Illumina , San Diego , CA ) . The sequencing was performed using a MiSeq sequencer ( Illumina ) at the Molecular and Cellular Biology ( MCB ) core at the ONPRC . The resulting DNA reads were aligned to the published genome sequence of MPXV-USA2003-039 ( GenBank accession # DQ11157 ) . Illumina sequence data were processed using a custom analysis pipeline written by B . N . B . This pipeline has been made available as a module for LabKey Server , an open-source platform for the management of scientific data [56] . The SequenceAnalysis module provides a web-based interface to initiate analyses , manage data , and view results . The source code behind this pipeline is available in a subversion repository ( https://hedgehog . fhcrc . org/tor/stedi/trunk/unsupportedModules/labModules/SequenceAnalysis ) . Raw reads were trimmed by sequence quality using Trimmomatic [57] and aligned against the reference genome using BWA-SW [58] . Single Nucleotide Polymorphisms ( SNPs ) between reads and the reference sequences were scored with scripts that utilized SAMtools , picard tools ( http://picard . sourceforge . net ) , and bioperl [59] , [60] . CHO cells were tranduced with either Ad-tTA ( 25 MOI ) or Ad-197 ( 20 MOI ) and Ad-tTA ( 5 MOI ) . At 24 h post transduction ( p . t . ) , the cells were washed with PBS , overlaid with DMEM ( Cys−/Met− ) , and incubated for 1 . 5 h . The cells were pulsed with 300 µCi/106 cells for 45 min and the label was chased for the indicated time intervals . CHO cells were washed with ice-cold PBS and lysed with ice-cold PBS-1% NP-40 buffer . Cell lysates were pre-cleared with agarose beads and immunoprecipitated with αFLAG Ab conjugated to agarose beads ( Sigma-Aldrich , St . Louis , MO ) . The samples were eluted from the beads with 50 mM NaOAc-0 . 15% SDS buffer ( 10 min , 98°C ) and treated with EndoH ( Roche Diagnostics , Indianapolis , IN ) or PNGase ( New England Biolabs , Ipswich , MA ) according to the manufacturer's protocols . The samples were separated on a 6% polyacrylamide gel . CHO cell lysates or immunoprecipitated samples were separated on 6% polyacrylamide gels and transferred onto Immobilon PVDF membranes ( EMD Millipore , Billerica , MA ) . The membrane was blocked with 5% skim milk in PBS-0 . 05% Tween 20 ( PBST ) buffer and blotted with αFLAG Ab ( Sigma-Aldrich , 1∶500 ) and secondary HRP-conjugated mouse TrueBlot Ab ( eBioScience , San Diego , CA ) diluted in 5% skim milk-PBST . The immunoblots were developed with SuperSignal West Pico Chemiluminescent Substrate kit ( Thermo Fisher Scientific , Rockford , IL ) . CHO cells grown in T75 flasks to 80% confluency were transduced with either Ad-tTA alone ( 25 MOI ) or Ad-197 ( 20 MOI ) and Ad-tTA ( 5 MOI ) or Ad-B22R ( 20 MOI ) and AdtTA ( 5 MOI ) . After 24 h incubation , the cells were washed twice with PBS and biotinylated using Pierce Cell Surface Protein Isolation kit ( Thermo Fisher Scientific , Rockford , IL ) according to the manufacturer's protocol . Biotinylated proteins were immunoprecipitated with NeutrAvidin agarose resin provided with the kit , separated on 6% PAGE gel , and blotted with αFLAG Ab . CHO cells were plated on glass coverslips in 12-well plates at 50% confluency . The next day the cells were transfected with 500 ng of indicated plasmids using lipofectamine 2000 ( Life Technologies ) according to the manufacturer's protocol . At 24 h p . t . , the cells were washed with ice-cold PBS , fixed with 4% paraformaldehyde , and permeabilized with 0 . 2% Triton X100 . The samples were blocked with 2% bovine serum albumin ( BSA ) -PBS ( P-BSA , pH 7 . 4 ) and stained with primary mouse αFLAG Ab ( 1∶1000 ) and secondary anti-mouse to Alexa Fluor 594 ( 1∶1000 , Life Technologies ) diluted in 2% P-BSA . The coverslips were mounted on slides in ProLong Gold antifade reagent with 4 , 6-diamidino-2-phenylindole ( DAPI; Life Technologies ) and analyzed with Leica TCS SP laser scanning microscope . MaMu-A*01 CM9 tetramer was kindly provided by Marcelo Kuroda ( Department of Immunology , Tulane National Primate Research Center ) . The tetramer was conjugated to Allophycocyanin ( APC ) using ProZyme PhycoPro GT5 APC kit ( Prozyme , Hayward , CA ) according to the manufacturer's protocol . Monkey CM9-peptide specific T cells recovered after co-incubation with Ad-197/Ad-tTA or Ad-tTA only infected CHO cells ( described above ) were incubated with the tetramer for 1 h at 37°C and stained with LIVE/DEAD Fixable Dead Cell Stain ( Life Technologies ) and Ab specific to CD95 ( clone DX2 , BD Biosciences ) , CD28 PE ( clone L293 , BD Biosciences ) , CD45 ( clone D058–1283 , BD Biosciences ) , CD8 ( clone SK1 , BD Biosciences ) , and CD3 ( clone SP34-2 , BD Biosciences ) for 30 minutes at room temperature . The cells were fixed with 2% paraformaldehyde and analyzed by flow cytometry as described above . Orthopox-specific enzyme-linked immunosorbent assay ( ELISA ) was performed as previously described [10] using whole-VACV lysate ( inactivated by pre-treatment with 3% H2O2 for 2 h ) . An internal positive control was included on each plate to normalize between plates and between assays performed on different days . Antibody titers were determined by log-log transformation of the linear portion of the curve , using 0 . 1 optical density units as the endpoint and performing conversion on final values . | We discovered that the largest gene in the genome of monkeypox viruses and several related viruses , including the virus causing smallpox , but not vaccine strains , encode a protein ( B22 ) that renders the cellular immune system non-responsive . A particularly novel aspect of this work is that B22 proteins directly disable cells of the immune system as compared to previously known molecular strategies that help viruses to hide from the immune system . We further show that monkeypox viruses containing this protein are much more virulent in non-human primates than viruses that lack B22 . Our observations suggest that B22 proteins contribute to monkeypox virulence and might have contributed to the severe disease manifestations of variola major virus . However , these data also suggest that B22 proteins could potentially be used to curb undesired immune responses such as autoimmunity or graft versus host disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology",
"and",
"life",
"sciences",
"immunology"
] | 2014 | T Cell Inactivation by Poxviral B22 Family Proteins Increases Viral Virulence |
S-adenosylmethionine ( SAM ) is a donor which provides the methyl groups for histone or nucleic acid modification and phosphatidylcholine production . SAM is hypothesized to link metabolism and chromatin modification , however , its role in acute gene regulation is poorly understood . We recently found that Caenorhabditis elegans with reduced SAM had deficiencies in H3K4 trimethylation ( H3K4me3 ) at pathogen-response genes , decreasing their expression and limiting pathogen resistance . We hypothesized that SAM may be generally required for stress-responsive transcription . Here , using genetic assays , we show that transcriptional responses to bacterial or xenotoxic stress fail in C . elegans with low SAM , but that expression of heat shock genes are unaffected . We also found that two H3K4 methyltransferases , set-2/SET1 and set-16/MLL , had differential responses to survival during stress . set-2/SET1 is specifically required in bacterial responses , whereas set-16/MLL is universally required . These results define a role for SAM in the acute stress-responsive gene expression . Finally , we find that modification of metabolic gene expression correlates with enhanced survival during stress .
Cellular functions are profoundly affected by metabolic state . For example , transcriptional regulation can be linked to metabolism through the modification of chromatin by methylation [1] . Using the methyl groups produced by the 1-carbon cycle ( 1CC ) and donated by S-adenosylmethionine ( SAM ) , histone methyltransferases ( HMTs ) can change the regulatory state of chromatin , promoting or limiting gene activity [2] . HMT activity can be controlled by recruitment of HMT containing complexes to specific genomic locations [2] . However , SAM availability may also affect histone methylation patterns [3] . SAM is produced by the 1-carbon cycle ( 1CC ) and levels can be affected by folate , methionine or choline levels or by other factors such as alcohol consumption [4] . Variations in SAM levels have been proposed to mediate transgenerational inheritance of epigenetic patterns or other gene regulatory events , however , a direct mechanistic connection has been difficult to establish [5] . Although SAM is necessary for all histone methylation events , in vivo studies have suggested that particular methylation marks are more sensitive to changes in SAM levels . For example , induced pluripotent stem cells ( iPSCs ) , murine liver , and C . elegans all show a decrease in H3K4me3 levels as SAM levels drop [6–9] . Furthermore , in budding yeast , a rise in SAM levels is followed by increases in H3K4me3 [10] . Furthermore , these linked changes in SAM levels and H3K4me3 also correlate with changes in cell-type-specific gene expression and differentiation in iPS cells [7] . Finally , Dai et al . have recently demonstrated that treatments with high and low methionine , which is the precursor for SAM , alters H3K4me3 peak width at genes in steady-state conditions in mouse liver and human cancer cells [11] . Thus , SAM levels are tightly linked to H3K4me3 dynamics . Trimethylation of H3K4 is a common modification occurring close to the start site of actively transcribed genes and is accomplished through the activity of the COMPASS complex [12] . The KTM2 family of HMTs serves as the enzymatic activity of COMPASS providing mono , di and trimethylated states [13] . In yeast , there is a single member , Set1 , whereas mammals can use one of seven enzymes , within subfamilies of SET1 , MLL ( Mixed lineage leukemia ) or THX ( Trithorax ) [12] . However , the relationship between H3K4me3 and transcription is complex , as it does not appear to be necessary for global gene expression in basal conditions [14] . In yeast , Set1 has an important role in limiting the expression of ribosomal genes during the response to diamide stress [15] suggesting that chromatin-modifying factors are especially critical when organisms experience stress . The H3K4 methyltransferase family in mammals appears to have overlapping as well as specialized functions in either specificity for mono , di or tri-methylation or through distinct roles in development [13] . However , clearly defined roles for each MT have been difficult to discern . C . elegans genome encodes a simplified KTM2 family containing three H3K4 methyltransferases , set-2/SET1 , set-16/MLL and ash-2/THX [16] . Interestingly , these methyltransferases have distinct developmental and tissue-specific biological functions . set-2/SET1 is broadly important for H3K4 trimethylation in embryos and the germline [17 , 18] and the intestine [8] . Also , loss or reduction of set-2/SET1 influences fertility across generations [19] , lifespan [20] and lipid accumulation [21] . ash-2 acts through the germline to affect lifespan and lipid accumulation in the intestine [20–22] . set-16/MLL , on the other hand , appears to be dispensable for H3K4me3 in the early embryo and germline ( Li , 2011 ) , while we found that it has a partial requirement in the adult intestine [8] . Thus , while H3K4me3 marks the start sites of actively transcribed genes , the methyltransferases producing it can have diverse and long-acting biological effects . Using a C . elegans model of low SAM , we previously found that transcriptional responses to a bacterial pathogen failed and these bacterial-response genes did not show the normal pattern of H3K4me3 close to the transcriptional start sites , [8] . We also found that the HMT set-16/MLL was required for full induction , whereas set-2/SET1 appeared dispensable [8] . We hypothesized that animals with low SAM might fail to transcriptionally respond to stress and that the HMTs may also have distinct roles in modulating stress responses . In our present study , we set out to compare induction of transcriptional responses and survival upon stress exposure between C . elegans with reduced SAM and animals with limited H3K4me3 function . Because distinct stresses may rely on different transcriptional activation mechanisms , we also compared whole-genome expression patterns in three stresses: pathogenic bacteria , xenotoxic and heat . We found that the induction of genes in the pathogen and xenotoxic stress response were diminished in low SAM , with concomitant reductions in survival in these animals . However , while pathogen and xenotoxic-stress genes were affected after both set-2/SET1 and set-16/MLL RNAi , set-16/MLL was uniquely required for survival in all three stresses . This suggests SAM and set-16 have essential functions in transcriptional responses to diverse stresses . Interestingly , induction of heat stress response genes , which are controlled primarily by promoter pausing of RNA Pol II [23] , occurs even in low SAM and after H3K4 methytransferase knockdown . While expression of canonical heat shock genes occurred in each of these conditions , sams-1 animals fared better than controls , set-2 animals survived at control levels and set-16 animals died rapidly . Thus , the expression of stress response genes and survival may not correlate in all cases . Finally , we find that in addition to stress-responsive genes , regulation of metabolic genes may be key to the survival of animals with deficient H3K4 methylation during stress .
Gene regulatory events can be controlled by histone methylation; however , it is not clear how levels of the methyl donor SAM may alter methylation patterns and gene expression in different physiological conditions ( Fig 1A ) . In C . elegans , we previously found animals with a mutation in the SAM synthase sams-1 , which have 50% of the SAM of wild-type animals [24] , had poor survival on the bacterial pathogen Pseudomonas aeruginosa [8] . SAM deficient animals failed to upregulate selected pathogen-response genes and had reduced global H3K4 trimethylation in intestinal nuclei as well as at specific pathogen-response genes [8] . We hypothesized this could represent a general failure of stress-responsive gene expression , as low SAM levels were unable to support rapid remodeling of H3K4 methylation as transcriptional needs changed . To test this model ( Fig 1B ) , we used RNAi to knockdown sams-1 or the H4K4me3 methyltransferases ( HMTs ) that use SAM , set-2 and set-16 , then exposed animals to three stresses: bacterial ( P . aeruginosa ) xenotoxic , or heat . For xenotoxic stress , we used R24 , an agent that robustly stimulates both immune and detoxification responses in C . elegans [25–27] . Next , we used whole genome RNA sequencing to determine which genes changed in each stress and assayed how they were affected by low SAM or depletion of the HMTs and selected genes with greater than two-fold change in any of the conditions with a false discovery rate ( FDR ) of <0 . 01 for further analysis . To determine if gene expression patterns were shared between control , sams-1 , set-2 and set-16 animals in response to P . aeruginosa , we mapped gene expression patterns with Venn diagrams for up ( Fig 1C–1F ) and down ( S1A–S1D Fig ) regulated genes . We found distinct , large scale gene expression changes with each stress , suggesting gene expression modules were specified by stress-specific mechanisms rather than by SAM or these H3K4 methyltransferases . We first examined gene expression changes between control and sams-1 ( RNAi ) in non-stressed conditions ( basal ) . SAM may contribute to PC production as well as to histone methylation ( S2A Fig ) and several lines of evidence from our previous study of gene allow us to distinguish indirect effects downstream of phospholipid methylation from other SAM-dependent events . In basal conditions , several hundred genes changed by more than two-fold after sams-1 ( RNAi ) ( S2B Fig , S1 Table ) , with significant overlap with our previous microarray results [8] . In that study , we found that most gene expression changes were linked to methylation-dependent PC production , as they were returned to wild-type levels when PC levels were rescued by dietary choline ( S2C Fig ) [8] . Stress-responsive genes activated downstream of PC were ( 1 ) expressed at low levels ( 2–5 fold ) and ( 2 ) expression was returned to wild-type levels by dietary choline , which rescued PC levels [8] . This is in contrast to activation of bacterial stress-responsive genes by P . aeruginosa , which was dynamic ( up to 250 fold ) and not responsive to choline [8] . Our observation that P . aeruginosa-responsive gene expression also depended on H3K4 methyltransferases suggests that stress responsive transcription might have a distinct requirement for SAM and H3K4 methyltransferases than in basal conditions . Finally , decreases in H3K4me3 in C . elegans intestinal nuclei were not rescued when choline returned PC to wild type levels , suggesting that SAM-dependent decreases in this histone methylation mark are not linked to indirect effects from PC . Both H3K4 methylation and PC production are significant consumers of SAM [3] ( S2A Fig ) . Recently Ye , et al . show that H3K4 tri-methylation can increase in Saccharomyces cerevisiae when PC production is blocked and SAM levels increase [10] . In agreement with these findings , we also observed that global H3K4me3 levels increase when the PC-producing methyltransferases pmt-2 was knocked down ( S2D Fig ) . Thus , in basal conditions , gene expression changes to compensate for decreases in PC are the predominant effect of sams-1 loss , with negligible effects due to other methylation pathways . Finally , modified H3K4 may exist in several methylation states [13] . Using immunostaining with antibodies to H3K4me1 and H3K4me2 , we found that levels did not decrease as they had with H3K4me3 ( S2E Fig; [8] ) , suggesting that the trimethylated state is most sensitive to SAM levels in adult C . elegans intestine . Next , we compared gene expression patterns in control , sams-1 , set-2 and set-16 RNAi animals during P . aeruginosa exposure . Control animals upregulated 651 genes more than two-fold in response to the bacterial stress ( Fig 2A and 2B , S2 Table ) with a high concordance to previous studies that identified P . aeruginosa-response genes [28 , 29] ( S2 Table ) . Heat maps comparing genes upregulated more than 2-fold with an FDR of < 0 . 01 show lower induction after sams-1 RNAi , with intermediate effects after set-2 or set-16 knockdown ( Fig 2A ) . Focusing on the top 20 expressed genes in control animals , we find a significant reduction in expression ( Fig 2B ) and finally , we find that few genes outside the pathogen response are induced after sams-1 RNAi ( Fig 2C ) . The transcriptional response to P . aeruginosa also includes downregulation of a small subset of genes [28] . Comparisons between control and sams-1 ( RNAi ) gene expression patterns show that a proportion of the top 20 downregulated genes in control animals fail to decrease after sams-1 RNAi ( S3A Fig ) and that only about 5 percent of these genes overlap ( S3B Fig ) . Thus , this whole genome data confirms our analysis of selected P . aeruginosa-responsive genes [8] and shows that SAM is essential for the broad transcriptional changes occurring during stress caused by a pathogenic bacteria , reducing both total numbers of regulated genes and their magnitude . To determine how sams-1 RNAi animals respond to a distinct stress , we treated control and sams-1 RNAi animals with R24 , a xenotoxic agent that induces both detoxification and innate immune defenses [25–27] . R24 was originally identified in a screen of 37 , 200 small molecules that utilized C . elegans as a heterologous host to identify new anti-infective compounds[30] . Interestingly , R24 protects nematodes from bacterial infection by boosting the transcription of innate immune defenses [26 , 27] . This molecule is also toxic to worms growing under normal laboratory conditions . Exposure to R24 strongly activates the transcription of cytochrome P450 and other detoxification genes , it shortens nematode lifespan and delays worm development [26 , 27] . We used R24 as a tool to compare gene expression changes in animals with low amounts of SAM . RNA-seq was performed on control and sams-1 ( RNAi ) after treatment with R24 . A set of genes was significantly upregulated by R24 in wild-type animals , which was consistent with published results [25–27] . Importantly , the induction of genes by R24 was significantly attenuated in sams-1 ( RNAi ) animals ( Fig 3A , S3 Table ) . The 20 most highly induced genes after R24 treatment included multiple cytochrome p450s as well as previously identified pathogen response genes ( Fig 3B ) [27] . Strip plots comparing levels in control and sams-1 ( RNAi ) animals show that each gene was markedly decreased ( Fig 3B ) . We also found that most of the genes induced in sams-1 animals were part of the response to R24 in control animals ( Fig 3C ) . R24 also induces the downregulation of a limited subset of genes [27] . Comparison between genes downregulated in control animal or after sams-1 RNAi shows that sams-1 RNAi also limits this downregulation ( S3C and S3D Fig ) . Thus , low SAM attenuates the transcriptional response to a xenotoxic agent , just as it does to bacterial stress-responsive gene expression induced by P . aeruginosa . Transcriptional response to bacteria or xenotoxic agents are predicted to follow a classic signal transduction pathway where the extracellular stimulus activates a cellular signaling pathway linked to individual transcription factors and upregulation of stress-specific gene expression [31] . However , other stress-responsive genes expression , such as the heat shock genes , are regulated differently . RNA Pol II is paused at promoters of many heat shock genes and released into its elongating form in response to heat [32] . To determine if low SAM had the same effects on heat shock-dependent transcription as the bacterial or xenotoxic stress , we performed whole genome RNA sequencing on control , sams-1 , set-2 and set-16 RNAi animals exposed to 37°C for one hour . We found that heat-shock genes such as hsp-16 . 41 , hsp-16 . 2 , hsp-16-11 , hsp-16 . 48 , hsp-16 . 49 and hsp-70 were strongly induced in control animals in these conditions ( Fig 3D–3F , S4 Table ) . In contrast to the bacterial or xenotoxic stress responses , comparison of control and sams-1 patterns for genes induced at least 2-fold shows similar patterns ( Fig 3D ) , suggesting that reduced SAM availability does not compromise the activation of heat-stress induced genes . Strip plots comparing expression of the top 20 genes activated in control animals compared to sams-1 shows that most of the highly expressed genes are similarly or more highly expressed after sams-1 RNAi ( Fig 3E ) . Finally , Venn diagrams confirm that the genes upregulated after heat shock in controls are also upregulated in sams-1 animals and that many genes ectopic to the heat shock response also increase ( Fig 3F ) , suggesting that additional gene expression are activated in sams-1 animals under heat stress . Control animals downregulated approximately 300 genes after heat shock; strikingly , nearly 2000 genes decreased in parallel sams-1 RNAi animals ( S3E and S3F Fig ) . Thus , while expression of heat shock genes seems to occur independently of SAM , other genes outside this classical response dramatically increase or decrease during heat shock in low SAM . Histone methyltransferases use SAM to modify specific histone residues , modifying the chromatin environment to provide distinct gene regulatory states . Yeast contain a single H3K4 HMT , which provides mono , di and trimethylated states [12] and functions within the COMPASS HMT complex . Mammals encode 7 H3K4 HMTs that have different specificity for methylation states [12] . However , the non-redundant biological functions have been difficult to discern . C . elegans contains 3 H3K4 HMTs that affect H4K4me3 , set-2/SET1 , set-16/MLL and ash-2/THX . These HMTs affect embryonic and germline development [17 , 18] and transgenerational inheritance through the germline [20 , 22] . In our previous studies , we investigated the roles of set-2/SET1 and set-16/MLL in the adult C . elegans intestine , which is a critical tissue in the pathogen response [33] . Because H3K4me3 has been associated with dynamically transcribed genes and our previous results showing an increase in H3K4 tri-methylation at promoters of selected P . aeruginosa responsive genes during infection [8] , we sought to determine if set-2/SET1 or set-16/MLL were downstream of SAM-dependent responses during the stress response . Neither set-2 or set-16 RNAi significantly affected gene expression in non-stressed conditions ( S4A and S4B Fig , S1 Table ) . In parallel with sams-1 ( RNAi ) RNA experiments ( Figs 1 and 2 ) , we exposed set-2 ( RNAi ) animals to P . aeruginosa , the xenotoxic agent R24 or heat stress , extracted RNA and performed RNA-sequencing . Unbiased hierarchical clustering analysis of all genes significantly upregulated by any of these stresses showed that set-2 and set-16 RNAi grouped within each stress , suggesting similar overall gene expression patterns ( Fig 2A ) . Next , we used the same computational tools as in the sams-1 analysis to compare expression patterns of P . aeruginosa response genes after set-2 RNAi . Interestingly , although heat maps show that activation of bacterial-stress responsive genes are diminished after set-2 RNAi , the effect is less severe than in sams-1 ( RNAi ) animals ( Fig 2A ) . Direct comparison of the 20 most highly expressed genes shows reduced expression of several genes in set-2 animals , in line with an intermediate effect between controls and sams-1 ( Figs 2A and 4A ) and Venn diagrams show the P . aeruginosa-responsive genes in set-2 animals were primarily included in the control response ( Fig 4B ) . Analysis of the genes two-fold downregulated in control animals shows that most of the genes were reduced at similar levels after set-2 RNAi and were part of the same transcriptional response ( S4A and S4B Fig ) . Taken together , this data suggests that set-2 RNAi mediates part of the response to P . aeruginosa in low SAM . Low SAM decreased the ability of C . elegans to respond transcriptionally to xenobiotic R24 ( Fig 3A–3C ) . Interestingly , knockdown of set-2 mirrored sams-1 RNAi in some respects , but not others . Like sams-1 , heat maps show that set-2 RNAi limited the number of genes upregulated by more than two-fold ( Fig 3A , S3 Table ) . As with the response to P . aeruginosa , set-2 RNAi animals show reductions in several of the top 20 R24-induced genes and have diminished genome-wide expression of R24 response genes ( Fig 4C ) . However , a set of 135 genes were induced in response to R24 in set-2 RNAi animals that were not upregulated in control samples ( Fig 4D ) , suggesting deregulation or expansion of the transcriptional response . Next , we examined the response to heat stress after set-2 RNAi and found similarities with the sams-1 response . First , many genes induced more than twofold are similar in set-2 RNAi and controls ( Fig 3D , S4 Table ) . Second , although most of the top 20 genes upregulated in control animals were expressed after set-2 RNAi at near normal levels ( Fig 4F ) . Strikingly , many genes ectopic to the control response were induced after set-2 RNAi ( Fig 4G ) . As in the upregulated gene sets , the downregulated genes in control animals in response to heat also decreased after set-2 RNAi . However , a large number of genes not downregulated in controls also decreased ( S5E and S5F Fig ) . Thus , set-2 appear to be important for full response to P . aeruginosa or R24 , but dispensable for genes induced by heat in control animals . Interestingly , knockdown of this H3K4 methyltransferase appears to deregulate or expand the stress response to both R24 and heat . Like set-2/SET1 , set-16/MLL is important for H3K4me3 in the C . elegans intestine [8] . In addition , we identified a critical role for set-16 in mediating P . aeruginosa-responsive gene regulation in our previous studies [8] . Therefore , we also compared bacterial , xenotoxic and heat stress induction in knockdown of set-16 to set-2 and sams-1 . Confirming our previous qPCR analysis of selected P . aeruginosa-responsive genes in set-16 RNAi animals , we found that set-16 was broadly important for expression of genes upregulated by bacterial stress ( Fig 2A ) . Many of the highest expressed genes in control animals during the P . aeruginosa response were diminished after set-16 ( RNAi ) ( Fig 5A , S3 Table ) . Finally , most of the genes upregulated by P . aeruginosa in set-16 RNAi animals were also upregulated in control samples ( Fig 5B ) . A proportion of the genes downregulated by P . aeruginosa in control animals were also downregulated in set-16 RNAi animals ( S6A and S6B Fig ) . Thus , sams-1 , set-2 , and set-16 all appear to have critical roles in regulating genes in response to bacterial stress in C . elegans . Responses to R24 in set-16 ( RNAi ) animals largely mirrored sams-1 knockdown but were distinct from set-2 . Both the number of expressed genes and the levels of the highest expressed genes were significantly decreased ( Figs 3A , 5C and 5D ) . Finally , the majority of genes that increased in set-16 ( RNAi ) animals also increased in controls ( Fig 5D ) . The majority of the genes downregulated by R24 in control animals were not similarly downregulated after set-16 RNAi ( S6C and S6D Fig ) As in sams-1 and set-2 knockdown , the top twenty expressed genes in control animals were expressed similarly in heat-shocked control and set-16 ( RNAi ) animals ( Fig 5E ) and set-16 ( RNAi ) animals deregulated or expanded heat-stressed induced gene expression patterns compared to controls ( Fig 5F ) . Like sams-1 ( RNAi ) or set-2 ( RNAi ) , set-16 animals upregulated and downregulated genes whose expression were not part of the response in control animals ( Fig 5I; S5E and S5F Fig ) . Taken together , our results suggest that low SAM , decreased set-2/SET1 and set-16/MLL activity all compromise bacterial stress-induced gene expression . Xenotoxic stress induced by R24 appeared to have a stronger requirement for sams-1 or set-16 , with many ectopic genes upregulated in the set-2 response to R24 . Finally , genes activated by heat shock appeared mostly unaffected by low SAM , decreased set-2/SET1 or set-16/MLL activity , suggesting that neither SAM or these HMTs are essential for their expression . However , each displayed a significant number of ectopic genes inductions in both up- and down-regulated gene sets . This suggests that complex regulatory interactions may lie downstream of SAM or the H3K4 tri-methylases during the heat shock response . These could include regulation of downstream transcription factors , methylation of other histone or DNA targets or methylation-independent activity of COMPASS complexes [12] . We found that sams-1 , set-2 , and set-16 were all required for the transcriptional response to bacterial stress , but differentially affected the response to the xenotoxic agent R24 . Moreover , although sams-1 , set-2 , and set-16 were not required for heat shock gene expression , a varied but significant number of ectopic genes increased or decreased expression when knockdown animals were subjected to heat shock . Next , we sought to determine how low SAM or H3K4 HMT knockdown affected survival during each stress response . Previously , we found that sams-1 ( lof ) animals had poor survival on P . aeruginosa , which was matched by attenuated expression of bacterial-stress responsive genes , and impairment H3K4me3 acquisition at bacterial-stress responsive genes after infection [8] . To determine if set-2 or set-16 ( RNAi ) animals shared this susceptibility to bacterial stress , we challenged control and knockdown animals with P . aeruginosa and determined survival rates . Concordant with the whole genome RNA sequencing data ( Figs 4A–4C and 5A–5C ) , we found that both set-2 and set-16 ( RNAi ) animals had significantly reduced survival on P . aeruginosa ( Fig 6A , S5 Table ) . Our whole-genome expression analysis showed that many of the genes upregulated in control animals were part of the well-described transcriptional response to P . aeruginosa [28] . However , we also noted other gene sets that could have essential survival functions . When using GO term analysis with the Gorilla website ( http://cbl-gorilla . cs . technion . ac . il/ ) , we found that 32% of recognized genes were not associated with a GO term ( S7 Fig ) . Thus , we built an annotation tool , Worm-Cat to categorize a more complete list of C . elegans genes and determine gene enrichment scores through Fisher's exact test . Worm-Cat allows assignment of broad physiological or molecular categories ( i . e . , stress response ) , and then subsequently identifies specific sub-categories ( pathogen , heavy metal , etc . ) ( See S6 Table for annotation table ) . If genes do not have a clear physiological function or are pleiotropic , molecular functions were used . We validated this tool by comparison with GO analysis of our previously published microarray data from sams-1 and sbp-1 ( RNAi ) [8] ( S7 Table ) . The most significant categories , such as stress response pathogen in sbp-1 ( RNAi ) and sams-1 ( RNAi ) upregulated genes or fatty acid metabolic genes in sbp-1 ( RNAi ) down , or sams-1 ( RNAi ) upregulated genes were identified by GO and by WORMCAT . While transcriptional regulation was identified by GO ontogeny for sams-1 ( RNAi ) upregulated genes , our tool showed a breakdown showing an enrichment for nuclear hormone receptors , providing additional specificity . This tool was also able to show enrichment for regulation of 1CC genes in sbp-1 downregulated genes , which we had previously noted [24] , but were not identified by GO ontogeny . Thus , this tool increases the depth and specificity of gene function in comparison to GO term enrichment . We used Worm-Cat to determine the major categories of genes that were changed in control , sams-1 , set-2 or set-16 RNAi animals , then determined which categories matched survival patterns . In large sets of regulated genes , super enrichment can be a valuable tool for assessing the role of genes with the largest difference in gene expression patterns [34] . Therefore , we determined enrichment scores for both 2 fold and 4 fold enriched genes ( Fig 6B–6D and S8 Table ) . Although stress-responsive genes were still enriched after sams-1 , set-2 or set-16 RNAi , both the enrichment score and gene number were reduced at both the Category 1 ( Fig 6B ) and more specific Category 2 levels ( Fig 6C , S8 Table ) . Surprisingly , metabolic genes were a significant fraction of genes upregulated in control animals and less enriched after sams-1 , set-2 or set-16 RNAi ( Fig 6B ) , correlating with poor survival of these animals . A breakdown of specific categories of metabolic genes showed that lipid metabolism was a significant category in control animals and that there were fewer genes and lower enrichment scores after sams-1 , set-2 or set-16 RNAi . Interestingly , fatty acid desaturases have been linked to response of C . elegans to P . aeruginosa [25] . In addition , 1CC genes are among those downregulated by C . elegans during P . aeruginosa infection ( [28] , see also S8 Table ) ) . Taken together , this suggests metabolic regulation may be an important part of these bacterial stress response . R24 is a xenotoxic agent that activates both immune and detoxification responses [25–27] . We found that transcriptional responses to R24 were distinct in sams-1 , set-2 , and set-16 animals , with genes ectopic to the control xenotoxic stress response increasing in set-2 RNAi animals ( Fig 3A–3C; Fig 4D–4F; Fig 5D–5F ) . To determine how sams-1 , set-2 and set-16 knockdown affected survival , we treated animals with R24 and monitored death rates . We found that knockdown of sams-1 , set-2 , and set-16 had differential susceptibility to R24-mediated toxicity . First , we found that concordant with the reduced expression of xenotoxic agent-response genes , sams-1 and set-16 had poor survival rates , with set-16 animals showing a particularly sharp decline ( Fig 6E , S9 Table ) . Knockdown of set-2 , however , did not decrease survival . Next , we used Worm-Cat to identify gene categories that might correlate with sensitivity to the xenotoxic agent in sams-1 and set-16 animals , or survival in the set-2 cohort . First , we noticed that as expected , stress-response genes were the most enriched category in control animals , with fewer genes in the sensitive sams-1 or set-16 animals ( Fig 6F , S10 Table ) . Breakdown of stress categories shows that R24-induced genes are enriched for cytochrome P450 genes and pathogen response genes ( Fig 6G ) , as expected for R24 [25–27] . However , the knockdowns most sensitive to R24 ( sams-1 and set-16 ) differed slightly in their stress response profiles . After RNAi of sams-1 , both CYP450 and pathogen response gene categories loose enrichment ( Fig 6G , S10 Table ) . However , set-16 lost enrichment only within the pathogen category ( Fig 6G , S10 Table ) , suggesting that genes within the pathogen response category may be important for survival on R24 . Supporting this notion , set-2 , which survived normally , lost CYP450 enrichment but retained genes in the pathogen response category ( Fig 6G , S10 Table ) . We also noted that metabolic categories were also limited in the sensitive strains . Metabolic genes , particularly in the lipid metabolism category were significantly enriched in Control and set-2 RNAi animals after R24 treatment , but enrichment scores failed significance after sams-1 or set-16 RNAi ( Fig 6F and 6H , S10 Table ) . This suggests that as in the bacterial stress response , rewiring metabolic genes correlates with stress survival . Notably , an RNAi screen for R24-dependent regulators of the innate immune response gene irg-4 identified multiple genes involved in fatty acid synthesis [25] . Finally , we found that genes deregulated in set-2 ( RNAi ) animals were enriched for genes activated by multiple stresses ( Fig 6F , S10 Table ) . However , since these genes have no other functional classification , their importance of this gene set is unclear . Transcription of heat shock genes in response to high temperature is controlled by shifting RNA pol II from a paused to the elongating form at heat shock response genes [32] . To provide a comparison to bacterial or xenotoxin-induced stress , we compared heat-shock responsive transcription in low SAM or after knockdown of the set-2/SET1 or set-16/MLL methyltransferases to transcriptional changes occurring after bacterial or xenotoxic stress responses . Strikingly , we found that many genes ectopic to the control heat shock response were activated or repressed after sams-1 , set-2 or set-16 RNAi ( Figs 3D–3F , 4H–4J and 5H–5J ) . Next , we performed survival assays to determine if these gene expression changes altered survival of these animals during stress . Unlike the bacterial stress response , sams-1 , set-2 , and set-16 all had distinct survival curves . First , sams-1 animals were markedly resistant during the first half of the assay , with the survival percentage at the assay midpoint more than twice that of control animals ( Fig 7A , S11 Table ) . The endpoint of the assay , however , was close to controls . Second , as in R24 assays , set-2 animals survived most similar to controls , although p values showed a significant difference ( Figs 6A and 7A , S11 Table ) . Finally , set-16 RNAi caused an extreme sensitivity to heat stress ( Fig 7A , S11 Table ) , similar to P . aeruginosa and R24 responses . To determine if categories of genes expressed in the sams-1 , set-2 or set-16 RNAi animals correlated with the differential survival , we used Worm-Cat to survey the enriched and super enriched heat responsive genes . As expected from our initial analysis , significant numbers of stress-responsive genes were enriched in the upregulated sams-1 , set-2 and set-16 RNAi cohorts ( Fig 7B and 7C , S12 Table ) . Genes in chromatin structure were also enriched in all but sams-1 RNAi animals ( Fig 7B and 7D , S12 Table ) . However , none of these category clusters correlated with survival . Next , we examined the categories enriched in the genes downregulated during the heat shock response ( Fig 7E–7G; S12 Table ) . While expression of canonical heat shock genes did not correlate with survival , we found enrichment in other functional categories . There were several categories of enriched genes ( transmembrane transport , proteolysis , and protein modification ) among the sams-1 , set-2 or set-16 RNAi animals ( Fig 7E , S12 Table ) . However , two categories correlated best with survival: metabolism and transcription factors . Strikingly , metabolism was only enriched as a downregulated category in sams-1 ( RNAi ) animals during heat shock , with the majority of these genes in anabolic pathways such as lipid and amino acid metabolism ( Fig 7E and 7F; S12 Table ) . This is distinct from PC-dependent effects from sams-1 in basal conditions , where fatty acid genes are activated ( see S7 Table , S12 Table ) . C . elegans contains a major expansion of nuclear hormone receptors , many of which are thought to regulate metabolic processes [35] . Intriguingly , we also observed reduced NHR gene expression sams-1 ( RNAi ) animals ( Fig 7G , S12 Table ) , concomitant with the loss of metabolic gene expression . Finally , we find that regulation of metabolic gene expression also correlates with survival in heat stress , as it did in our bacterial or xenotoxic stress assays , suggesting metabolic flexibility may be a common effector in stress response survival ( S8 Fig ) .
Metabolites that contribute to cellular regulatory functions , such as the methyl donor SAM , could be predicted to have broad effects on transcription . Indeed , SAM has been proposed as a link between nutrition and regulation of the 1CC to transgenerational epigenetic effects [5] . However , work from several labs across multiple systems has shown that SAM has surprisingly specific effects on histone methylation , reducing or increasing H3K4me3 as levels fall or rise [6–9] . Since H3K4me3 is tightly associated with start sites of actively transcribed genes , this suggests SAM may also have a critical role in acute gene regulatory events . Recently , the Locasale lab has shown that H3K4me3 peak breadth is sensitive to methionine levels in mouse liver and human cancer cells , strengthening the connections between SAM and H3K4me3 in vivo [11] . In this study , we have defined a role for SAM in the regulation of two stress responses , bacterial and xenotoxic stress , and found that it is necessary for induction of specific response genes , as well as for survival . This link between 1-carbon metabolism and stress responses has important implications for how organisms can respond to stress when metabolically challenged . Interestingly , while heat shock genes were expressed independently of SAM , many other genes had altered expression patterns in heat shocked sams-1 ( RNAi ) animals which survived better than wild type . This suggests multiple independently regulated modules can contribute to survival . While the expression of heat-shock response genes at control levels in sams-1 ( RNAi ) animals suggest these are regulated independently of SAM , the ectopically regulated genes could respond to SAM from direct or indirect mechanisms . For example , Labbadia and Morimoto have recently shown that in C . elegans , non-cell autonomous mechanisms linked to repressive H3K27me3 limit stress responses when reproduction starts [36] . Thus , this ectopic gene activation in sams-1 ( RNAi ) animals could result from changes in repressive methylation on other histones , on DNA or through other indirect effects . Our Worm-Cat annotation tool has shown that downregulation of two classes of genes correlates with survival of sams-1 RNAi animals under heat shock: metabolic genes ( amino acid , lipid and beta-oxidation ) and nuclear hormone receptors . We hypothesize that downregulating these categories pauses anabolic processes and allows a survival advantage for sams-1 animals as they respond to heat stress . Interestingly , fatty acid synthesis genes are upregulated after sams-1 RNAi in basal conditions , as changes in PC levels activate the lipogenic transcription factor SBP-1 [24] . As in our study of low SAM on C . elegans on normal laboratory diet of E . coli or P . aeruginosa where we found differential effects on activation or repression of pathogen response genes [8] , this suggest the effects of low SAM may differ in distinct stress or nutritional conditions . Future metabolomic studies will be important for how these changes in metabolic gene expression are linked to survival during heat stress . Nuclear hormone receptors are common regulators of metabolic genes [35] , and while direct relationships between these nuclear hormone receptors and the metabolic genes identified in our study are not yet discernable , it is intriguing that both classes of these genes are downregulated in the surviving animals . Thus , low SAM may have both direct and indirect effects that influence gene expression and survival during stress responses . SAM is utilized by HMTs such as set-2/SET1 or set-16/MLL to produce methyl marks such as H3K4me3 . Intriguingly , KTM2s are among the most sensitive HMTs to SAM levels [5] . SET1 is the single H3K4me3 in yeast , and thus essential for all H3K4 methylation [12] . Neither set1 or H3K4 tri-methylation are essential for viability under standard conditions [14] . However , set1 appears to function to limit the expression of ribosomal genes during the response to diamide [15] . The mammalian methyltransferase family is complex with seven H3K4 methyltransferases that differ in specificity for mono , di or trimethylation [12] . However , it has been difficult to assign specific biological functions . In C . elegans where the KTM2 family is simpler , we have found that set-2 or set-16 RNAi mirrors some of the effects of low SAM , reducing transcriptional responses to multiple stresses . However , set-2/SET1 and set-16/MLL appeared to have distinct functional profiles during these stress responses . set-2/SET1 RNAi is similar to low SAM in response to bacterial stress . Some SAM-dependent P . aeruginosa responsive genes are also limited in expression after set-2 RNAi , and set-2 animals survive poorly . Our previous analysis of the P . aeruginosa response in set-2 and set-16 animals suggested that set-2 may have a more limited role [8] . The present whole genome study also bears out an essential role for set-16 in P . aeruginosa-responsive transcription , but notably , set-2 animals are also survived poorly on P . aeruginosa , suggesting that critical genes are limited in both cases . However , set-2 appears less critical for some genes in the detoxification response to R24 . Survival is close to wild-type , and intriguingly , metabolic genes related to lipid synthesis are upregulated , distinct from the control response . Like the response to low SAM during heat shock , set-2 RNAi did not limit expression of genes induced by heat shock in control animals while many genes were de-repressed or ectopically regulated outside the response in controls . As with R24 , set-2 RNAi animals survived similar to controls during heat stress . Therefore changes in gene expression did not impact these biological functions . set-2 also has intriguing functions during lifespan regulation in C . elegans . Greer et al . showed transgenerational effects on lifespan in set-2 mutants , and another study from the Brunet lab suggested that set-2 and another H3K4 HMT ( ash-2 ) linked lipid synthesis and lifespan regulation [21] . They found that ash-2 was important for non-cell autonomous germline to intestine regulation of these processes [21] . However , the role of set-2 in direct regulation was less clear . Notably , in our study , although lipid biosynthetic genes were not changed in set-2 animals at the L4/young adult time point , many of these genes did increase upon R24 treatment . Taken together , this suggests set-2 may impact the regulation of lipid synthesis genes at different points during C . elegans lifespan or during specific stress responses . set-16/MLL , on the other hand , was essential for survival in each of the stresses we tested . Transcriptional responses to bacterial stress and R24 were attenuated , similar to low SAM . sams-1 and set-16 RNAi animals both survived poorly on P . aeruginosa and following exposure to R24 . However , the set-16 RNAi animals were particularly sensitive to R24 . Interestingly , set-16 animals were more deficient in activating pathogen response than CYP in response to R24 suggesting that pathogen response genes may be essential for survival . Like sams-1 and set-2 RNAi , heat stress of set-16 ( RNAi ) animals produced similar activation to control in the top 20 genes , in addition to ectopic activation or derepression of many other genes . However , this did not enhance survival . Taken together , this suggests that set-16 has a distinct role in survival during diverse stress responses . During a stress response , many genes must be coordinately regulated downstream of specific signaling pathways . For example , pathogenic stress may be sensed by activation of Toll-like receptors in mammals and Drosophila [37] or by translational attenuation in C . elegans [38] . These signals are carried through stress-specific transcription factors that activate protective genes . Along with these direct regulatory pathways , the chromatin environment must be permissive . It is intriguing that a metabolic pathway producing the methyl donor SAM and the H3K4 methyltransferases set-2 and set-16 are critical to enable transcriptional responses to acute stress . This suggests that 1CC status could influence how cells or organisms could respond to outside insults . The Halsted lab , using a micropig model of alcoholic fatty liver disease , has found that dietary limitation of methyl donors markedly decreases the time for development of liver injury [39 , 40] , thus , we suggest that low SAM could exacerbate disease progressing by limiting the ability of a tissue to respond to additional stress . This could increase the severity or the progression of a disease by limiting cellular defensive responses . Finally , other metabolites such as Acetyl CoA and NAD+ also influence gene regulation [3] . By having multiple metabolic pathways influencing histone modification and gene regulation , cells might finely tune transcription to diverse nutritional signals providing templates for specific metabolic states .
C . elegans ( N2 ) were cultured using standard laboratory conditions on E . coli OP50 . Adults were bleached onto RNAi plates for control ( L4440 ) , sams-1 , set-2 or set-16 and allowed to develop to the L4 to young adult transition before stresses were applied . For bacterial stress RNA preparations , nematodes were placed on E . coli or P . aeruginosa plates for 6 hours . For xenotoxic stress applications animals were placed on DMSO or 70 uM R24 plates for 18 hours . For heat stress applications , animals were raised at 15°C from hatching then at the L4/young adult transition replicate plates were placed at 15°C or 37°C for 1 hour . After each stress , animals were washed off the plates with S-basal , then pellets frozen at -80°C . RNA was prepared as in Ding , et al . 2015 [8] . For survival assays , animals remained on plates until all nematodes were dead . Exposure to Pseudomonas or R24 continued for the life of the animals . Exposure to heat occurred for 120 minutes , then animals were kept at 20°C for the remainder of the assay . Dead animals were identified by gentle prodding and removed each day . Kaplan-Meir curves were generated with OASIS [41] . Immunofluorescence was performed as in Ding et al . for H3K4me3 staining . For mono or di methyl staining , animals were fixed in 1% paraformaldehyde and permeabilized in cold 100% methanol before proceeding with the remainder of the protocol used in Ding , et al . 2015 . Antibodies used were: Tri-Methyl-Histone H3 ( Lys4 ) Rabbit mAb #9751 ( Cell Signaling ) , Abcam Anti-Histone H3 ( di methyl K4 ) antibody—ChIP Grade ( ab7766 ) ( Abcam ) and Anti-Histone H3 ( mono methyl K4 ) antibody—ChIP Grade ( ab889 ) ( Abcam ) . RNA for deep sequencing was purified by Qiagen RNA easy . Duplicate samples were sent for library construction and sequencing at BGI ( China ) . Raw sequencing reads were processed using an in-house RNA-Seq data processing software Dolphin at University of Massachusetts Medical School . The raw read pairs first were aligned to C . elegans reference genome with ws245 annotation . The RSEM method was used to quantify the expression levels of genes ( Li & Dewey , 2011 , PMID: 21816040 ) . All RNA sequencing data is available at the Gene expression ominibus , accession numbers , GSE121511 , GSE121509 , GSE121510 . Graphing for scatter and strip plots , Venn diagrams and bubble charts was done in R . The ontogeny category tool ( Worm-Cat ) consists of three parts . First , over 16 , 000 C . elegans genes were annotated; first by physiological role , then by molecular function . Categories contain up to three levels , for example , Proteolysis Proteasome: E3: F-box could appear as Proteolysis Proteasome in the broad Category 1 or as Proteolysis Proteasome: E3 or Proteolysis Proteasome: E3: F-box in the more specific categories 2 and 3 . Genes with broad physiological functions ( e . g . , ama-1 , RNA polymerase II large subunit ) were retained in molecular function categories . Phenotype data from alleles or RNAi were used to annotate physiological role if corroborated in two or more different assays . In addition , genes with no other function whose expression was changed by at least two of these stresses ( Methylmercury , tunicamycin , rotenone , cadmium , ethanol , D-glucose ) were placed in the category: Stress response: regulated by multiple stresses . Annotations were applied to genes regulated in each condition , then statistical significance of category enrichment determined by Fisher’s exact test with a p-value of < 0 . 05 used to determine significance . | Animals respond to stress by activating suites of protective genes . A specific metabolite , S-adenosylmethionine ( SAM ) , influences how these genes are activated in a variety of stress conditions . SAM is produced by the 1-carbon cycle and is the major donor for methylation reactions . Thus , SAM is used in the modification of histones , DNA , RNA and production of phospholipids . Here , we show that C . elegans with low SAM have reduced responses to a bacterial and toxic stress , but respond normally to heat stress . We also analyzed how animals that have reduced activity in some of the enzymes that use SAM to modify histones might respond to stress . One enzyme , SET-2 , was needed only for survival in bacterial stress , whereas the other related enzyme , SET-16 , was universally required . The availability of SAM may be affected by diets low in choline or methionine , alcohol or diseases such as cystic fibrosis . Thus , low SAM availability may leave organisms less able to respond to additional stress , which could exacerbate tissue injury or disease progression . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"classical",
"mechanics",
"pathology",
"and",
"laboratory",
"medicine",
"caenorhabditis",
"gene",
"regulation",
"pathogens",
"enzymes",
"microbiology",
"enzymology",
"mechanical",
"stress",
"animals",
"pseudomonas",
"aeruginosa",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"experimental",
"organism",
"systems",
"epigenetics",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"pseudomonas",
"transcriptional",
"control",
"genetic",
"interference",
"animal",
"studies",
"medical",
"microbiology",
"thermal",
"stresses",
"proteins",
"gene",
"expression",
"microbial",
"pathogens",
"methyltransferases",
"physics",
"biochemistry",
"rna",
"eukaryota",
"nucleic",
"acids",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] | 2018 | Stress-responsive and metabolic gene regulation are altered in low S-adenosylmethionine |
The human cytomegalovirus ( hCMV ) major immediate-early 1 protein ( IE1 ) is best known for activating transcription to facilitate viral replication . Here we present transcriptome data indicating that IE1 is as significant a repressor as it is an activator of host gene expression . Human cells induced to express IE1 exhibit global repression of IL6- and oncostatin M-responsive STAT3 target genes . This repression is followed by STAT1 phosphorylation and activation of STAT1 target genes normally induced by IFNγ . The observed repression and subsequent activation are both mediated through the same region ( amino acids 410 to 445 ) in the C-terminal domain of IE1 , and this region serves as a binding site for STAT3 . Depletion of STAT3 phenocopies the STAT1-dependent IFNγ-like response to IE1 . In contrast , depletion of the IL6 receptor ( IL6ST ) or the STAT kinase JAK1 prevents this response . Accordingly , treatment with IL6 leads to prolonged STAT1 instead of STAT3 activation in wild-type IE1 expressing cells , but not in cells expressing a mutant protein ( IE1dl410-420 ) deficient for STAT3 binding . A very similar STAT1-directed response to IL6 is also present in cells infected with a wild-type or revertant hCMV , but not an IE1dl410-420 mutant virus , and this response results in restricted viral replication . We conclude that IE1 is sufficient and necessary to rewire upstream IL6-type to downstream IFNγ-like signaling , two pathways linked to opposing actions , resulting in repressed STAT3- and activated STAT1-responsive genes . These findings relate transcriptional repressor and activator functions of IE1 and suggest unexpected outcomes relevant to viral pathogenesis in response to cytokines or growth factors that signal through the IL6ST-JAK1-STAT3 axis in hCMV-infected cells . Our results also reveal that IE1 , a protein considered to be a key activator of the hCMV productive cycle , has an unanticipated role in tempering viral replication .
Janus kinase-signal transducer and activator of transcription ( JAK-STAT ) signaling pathways are the principal means by which responses to dozens of cytokines , growth factors and other extracellular molecules are transduced from the cell surface to the nucleus . Although all JAK-STAT pathways share the same design principle , they involve distinct sets of ligands that engage different receptor and effector components to activate groups of genes which only partly overlap [1 , 2] . For interleukin ( IL ) 6 family cytokines , including IL6 and oncostatin M ( OSM ) , JAK-STAT pathway activation begins with ligand binding to specific receptors , such as the IL6 receptor ( IL6Rα or IL6R ) and the OSM receptor , respectively . The ligand-receptor interaction is followed by dimerization of the IL6 signal transducer ( IL6Rβ , GP130 or IL6ST ) subunits common to all IL6 family cytokine receptors . IL6ST is constitutively associated with several JAK family tyrosine kinases ( JAK1 , JAK2 and TYK2 ) of which JAK1 seems to be the most important for signaling in response to IL6 [3 , 4] . Upon receptor activation , JAK1 is phosphorylated and the activated kinase subsequently phosphorylates tyrosine residues in the cytoplasmic tail of IL6ST . These phosphotyrosines serve as docking sites for the src homology 2 ( SH2 ) domain of cytoplasmic STAT3 . Following recruitment to the receptor , STAT3 is phosphorylated on a single tyrosine residue ( Y705 ) by JAK1 or other kinases . Y705 phosphorylation is required for the formation of functional STAT3 dimers ( typically homodimers ) through reciprocal SH2-phosphotyrosyl interactions . The active pSTAT3 dimers subsequently dissociate from the receptor and accumulate in the nucleus , most likely coordinate with their ability to bind DNA [5] . DNA binding occurs rather sequence-specifically , resulting in transcriptional activation of select target genes involved in diverse processes including cell survival and proliferation [1 , 6 , 7] . One of the pSTAT3 target genes encodes the suppressor of cytokine signaling 3 ( SOCS3 ) which forms part of a negative feedback circuit by inhibiting IL6 signaling [8 , 9] . Another group of cytokines , the interferons ( IFNs ) , are distinct from the IL6-type cytokines but also trigger signaling through JAK-STAT pathways . For type I IFNs , including IFNα and IFNβ , canonical signaling occurs through the IFNα/β receptor subunits ( IFNAR1 and IFNAR2 ) , JAK1 and TYK2 , and a trimeric complex of tyrosine-phosphorylated STAT1 and STAT2 with IFN regulatory factor 9 ( IRF9 ) . This complex is also referred to as IFN-stimulated gene factor 3 ( ISGF3 ) and activates transcription of numerous genes many of which encode anti-viral products [10 , 11] . For IFNγ , the only type II IFN , signaling is typically mediated via the IFNγ receptor ( IFNGR ) subunits 1 and 2 , JAK1 and JAK2 , and tyrosine ( Y701 ) -phosphorylated STAT1 ( pSTAT1 ) homodimers . Like other STAT proteins , STAT1 also undergoes serine ( S727 ) phosphorylation adding to its potency as a transcriptional activator . By triggering prolonged STAT1 activation , type II IFN signaling results in the induction of numerous genes broadly defined as immune-modulatory [12 , 13] . The pathways responsive to IL6-type cytokines or IFNγ share important intracellular signaling molecules , but have been linked to opposing actions . STAT3 generally promotes cell survival and proliferation , may counteract inflammation and induces immune tolerance . In contrast , STAT1 tends to promote apoptosis , inhibits proliferation and favors innate or adaptive immune responses . Due to cross-regulation between the two pathways , perturbations in the levels or activities of STAT1 and STAT3 may redirect cytokine signals with unexpected outcomes [14–17] . Many viruses target components of JAK-STAT pathways including STAT1 , STAT2 and STAT3 . STAT1 and STAT2 usually act anti-viral due to their essential roles in IFN signaling [10 , 12] . Accordingly , most viruses antagonize STAT1 or STAT2 [18 , 19] , although viral activation and annexation of STAT1 has also been reported [20–28] . For STAT3 , the role in viral infections appears to be more complex . Thus , viruses either positively or negatively affect the expression or activity of STAT3 [29–36] . Human cytomegalovirus ( hCMV ) , one of eight human herpesviruses , is a very widespread opportunistic pathogen . To accomplish efficient replication and lasting persistence , hCMV seems to tweak most , if not all , host cell signaling pathways [37–40] . The 72-kDa ( 491 amino-acid ) immediate-early 1 protein ( IE1 ) has emerged as hCMV’s key modulator of JAK-STAT signaling [41–46] . Following infection of permissive cells , IE1 is among the very first and most abundant gene products produced de novo from the hCMV genome . The viral protein accumulates in the host cell nucleus and sets the stage for efficient hCMV early gene expression and subsequent viral replication [47–51] . The first hint suggesting IE1 may impact JAK-STAT pathways came from our finding that the protein confers increased type I IFN resistance to hCMV without negatively affecting IFN expression [52] . This phenotype was partly attributed to nuclear complex formation between IE1 and STAT2 depending on amino acids 373 to 445 [53] or 421 to 475 [54] in the viral protein’s C-terminal domain ( amino acids 373 to 491 ) . This domain is thought to be structurally largely disordered and contains four patches with highly biased amino acid composition: three acidic ‘domains’ ( AD1-AD3 ) and one serine/proline-rich stretch ( S/P ) [41 , 53 , 55] . The sequences downstream from the STAT2 interaction site in the C-terminal domain of IE1 feature a small ubiquitin-like modifier ( SUMO ) conjugation motif ( amino acids 449–452 ) [56–58] and a chromatin tethering domain ( CTD , amino acids 476–491 ) [59–61] which mediate binding to SUMO1 and to the acidic pocket formed by histones H2A-H2B on the nucleosome surface [62] , respectively . SUMOylation of IE1 may negatively regulate STAT2 binding [54] and positively affect hCMV replication [58] . IE1-STAT2 interaction causes diminished sequence-specific DNA binding by ISGF3 and inhibited type I ISG activation in the presence of IFNα or IFNβ [52–54 , 63] . The viral protein’s ability to inhibit type I ISG induction via STAT2 interaction is believed to be important , because it contributes to efficient hCMV replication [53 , 54] and appears to be conserved across IE1 homologs of the β-herpesvirus subfamily [64] . Besides functioning as an antagonist of type I IFN signaling , IE1 can also act as an agonist of type II IFN signaling . Following expression under conditions mimicking the situation during hCMV infection , IE1 elicited a host transcriptional response dominated by the up-regulation of genes normally induced by IFNγ . The IE1-dependent gene activation proved to be independent of IFNγ and other IFNs , yet required the Y701-phosphorylated form of STAT1 . Accordingly , IE1 induced Y701 and S727 phosphorylation , nuclear accumulation and binding of STAT1 to type II ISG promoters [21] . Whether IE1 binds to STAT1 directly or only indirectly ( via STAT heterodimers ) has not been resolved . Finally , STAT3 was shown to physically interact with IE1 , most likely via direct binding . The functional consequences of this interaction include STAT3 nuclear accumulation , disruption of IL6-induced STAT3 Y705 phosphorylation and inhibition of STAT3 binding to the SOCS3 promoter . These events are followed by diminished STAT3-dependent SOCS3 induction upon hCMV infection or IE1 expression [30] adding to the emerging evidence for transcriptional repression by the viral protein . However , IE1 has mostly been recognized as an activator of cellular and viral gene expression [42 , 65] and , to the best of our knowledge , no genome-wide analysis of human genes repressed by the viral protein has been pursued . Here we show , based on genome-wide transcriptome data , that IE1 is as much a repressor as it is an activator of human gene expression . We further demonstrate that a single motif ( amino acids 410 to 420 ) in the C-terminal domain of IE1 links the viral protein’s repressor and activator functions by rewiring upstream IL6-type to downstream IFNγ-like signaling resulting in repressed STAT3- and activated STAT1-responsive genes . Finally , the diversion of STAT3/STAT1 signaling attenuates viral replication revealing an unanticipated temperance activity in IE1 .
We previously reported on an Affymetrix GeneChip analysis of human transcripts undergoing up-regulation in MRC-5 cells transduced to express doxycycline ( dox ) -inducible IE1 ( TetR-IE1 cells ) . Expression from the preponderant majority ( >98% ) of genes represented on the GeneChips was not significantly affected by IE1 . However , a set of genes were specifically and reproducibly up-regulated by the viral protein [21] . Upon further inspection of the GeneChip data , we noticed that the IE1-dependent changes in the human transcriptome were not biased towards activation . Instead , the numbers of genes significantly up- or down-regulated by IE1 were roughly the same ( 410 up-regulated and 436 down-regulated probe sets ) . Notably , at 24 h post IE1 induction there were fewer ( <42% ) up-regulated compared to down-regulated genes , whereas after 72 h of IE1 expression the up-regulated ( >52% ) slightly outbalanced the down-regulated genes ( Fig 1 and S1 Data ) . These data indicate that IE1 is not only an activator , but also a significant repressor of host gene transcription . Our previous study has shown that the genes activated by IE1 in the TetR-IE1 cell model are largely responsive to pSTAT1 and IFNγ [21] . To identify upstream regulators common to genes repressed by IE1 , we used Ingenuity Pathway Analysis . Among probe sets exhibiting negative fold changes of ≥1 . 5 , this analysis identified highly significant associations with STAT3 ( p-value of overlap with reference set = 2 . 5×10−13 ) , OSM ( p = 6 . 2×10−9 ) , IL6 ( p = 1 . 2×10−7 ) and other components of STAT3-dependent signaling pathways including cytokines ( e . g . , leukemia inhibitory factor [LIF] , p = 8 . 2×10−8 , and granulocyte colony-stimulating factor [GCSF] , p = 8 . 4×10−7 ) , growth factors ( e . g . , hepatocyte growth factor , p = 2 . 2×10−5 , and epidermal growth factor , p = 2 . 6×10−4 ) and receptors ( e . g . , IL10 receptor α , p = 1 . 2×10−6 , and IL6ST , p = 1 . 2×10−4 ) . Consistently , 50 genes known to be positively regulated by STAT3 , IL6 or/and OSM were identified among the IE1-repressed genes based on Ingenuity Pathway Analysis ( Table 1 ) . Moreover , the majority of genes most significantly repressed by IE1 are known STAT3 targets ( Table 2 ) . Out of the genes identified to be repressed by IE1 , we selected six ( C4A , CHL1 , CXCL12 , IFI16 , RASL11A and SOCS3 ) for validation by reverse transcriptase quantitative PCR ( RT-qPCR ) . The genes were selected to reflect the full range of repression magnitudes and kinetics measured by GeneChip analysis . The RT-qPCR approach confirmed IE1-dependent down-regulation of all tested genes ( Fig 2A ) . Many genes identified to be repressed by IE1 , such as CXCL12 , IFI16 and SOCS3 , are known targets of STAT3 or its upstream activators including IL6 and OSM ( Tables 1 and 2 ) . However , to our knowledge , C4A , CHL1 and RASL11A have not been previously linked to activation by STAT3 , IL6 or OSM . We therefore examined the effects of IL6 and OSM treatment on the mRNA levels of our select set of IE1-repressed genes . Since fibroblasts do not express sufficient levels of the IL6 receptor α subunit ( IL6R ) to mount a robust response , IL6 was used in combination with a soluble form of IL6R . The results demonstrate that all tested genes are activated by both IL6 and OSM , although to varying degrees ( Fig 2B and 2C ) . To investigate whether these genes are also STAT3-responsive , STAT3 was silenced with two different siRNA duplexes . Both siRNAs were equally efficient in knocking-down STAT3 expression as confirmed by immunoblotting ( Fig 2D , left panel ) . Following depletion of STAT3 with either of the two siRNAs , all six tested genes exhibited reduced levels of expression ( Fig 2D , right panel ) . Likewise , STAT3 knock-down using two different dox-inducible shRNA constructs demonstrated STAT3-responsive expression for 12 out of 14 tested genes ( S1 Fig ) . Finally , we tested the effects of a mutant STAT3 protein ( STAT3α_Y705F ) , which is expressed to similar levels as the wild-type protein and resistant to Y705 but not S727 phosphorylation ( Fig 2E , left panel ) . This mutant protein is known to act in a trans-dominant negative fashion on expression of STAT3-responsive genes [66] . Accordingly , overexpression of STAT3α_Y705F resulted in reduced levels of all tested genes repressed by IE1 ( Fig 2E , right panel ) . These findings support the conclusion that most genes found to be down-regulated by IE1 in our system are IL6- and OSM-responsive pSTAT3 target genes , including genes not previously linked to this pathway . Next , we set out to map the physical determinants of STAT3-directed repression in IE1 and to relate them to other known activities of the viral protein . Previous work by us and Huh et al . has narrowed down STAT2 interaction to a region between amino acids 373 and 445 or 421 and 475 , respectively , in the C-terminal quarter of IE1 [53 , 54] suggesting that STAT3 might also bind to this part of the viral protein . The four low complexity stretches ( AD1 , S/P , AD2 and AD3 ) , the sequences ( including the SUMOylation motif ) linking these stretches and the CTD were individually deleted resulting in a set of eight mutant proteins spanning the entire IE1 C-terminal region between amino acids 373 and 491 ( Fig 3A ) . Subsequently , MRC-5 cells were transduced with lentiviruses expressing the wild-type or mutant IE1 proteins to generate a correspondent set of dox-inducible cell lines . Following induction , the steady-state IE1 levels in each mutant cell line were comparable to those of TetR-IE1 cells expressing the full-length viral protein ( Fig 3B ) . The IE1 mutants lacking an intact VKSE motif ( IE1dl446-450 and dl451-475 ) failed to undergo SUMOylation , as expected . IE1dl476-491 was not SUMOylated either , suggesting that the CTD is required in addition to the VKSE motif for this posttranslational modification ( Fig 3C ) . All mutant proteins displayed nuclear localization undistinguishable from wild-type IE1 , as determined by immunofluorescence microscopy ( Fig 4A ) . When testing for nuclear accumulation of STAT3 , cells expressing IE1dl373-386 , dl387-394 , dl395-409 , dl446-450 , dl451-475 or dl476-491 closely resembled wild-type expressing TetR-IE1 cells in exhibiting predominantly nuclear diffuse STAT3 staining ( >60% of cells ) or , less frequently , a balanced distribution of STAT3 between the nucleus and cytoplasm ( <40% of cells ) . By contrast , in most cells expressing IE1dl410-420 , dl421-445 or no IE1 , STAT3 was either evenly distributed across the nucleus and cytoplasm ( >60% of cells ) or predominantly present in the cytoplasm ( <40% of cells ) , but rarely if at all enriched in the nucleus ( Fig 4A and 4B ) . The immunofluorescence results were independently confirmed by subcellular fractionation analysis showing reduced STAT3 ( but not STAT2 ) nuclear accumulation in cells expressing IE1dl410-420 or no IE1 compared to wild-type IE1 expressing or IL6-treated cells ( Fig 4C ) . In agreement with the subcellular localization analyses , IE1 amino acids 405 to 491 were sufficient ( S2 Fig ) and amino acids 410 to 420 were required ( Fig 4D ) for physical interaction with STAT3 when mutants were compared to the full-length viral protein in co-immunoprecipitation assays ( Fig 4D ) . Unlike wild-type IE1 , the mutant protein also proved to be unable to interfere with STAT3 binding to SOCS3 promoter sequences , as analyzed by chromatin immunoprecipitation ( ChIP ) assay ( Fig 4E ) . Finally , repressed expression of the STAT3-responsive CXCL12 and SOCS3 genes was observed with the wild-type and all tested mutant proteins except for IE1dl410-420 and dl421-445 ( Fig 4F ) . Collectively , these results indicate that the STAT3-related activities of IE1 do not involve SUMOylation or nucleosome binding . Instead , residues in a region of IE1 that links the S/P and AD2 low complexity stretches ( amino acids 410 to 420 ) as well as residues within the AD2 motif ( amino acids 421–445 ) are required for interaction with STAT3 and subsequent repression of STAT3-responsive genes . Our previous work has demonstrated that IE1 triggers a type II IFN-like response via a mechanism that is dependent on tyrosine ( Y701 ) phosphorylation and enhanced by serine ( S727 ) phosphorylation of STAT1 induced by the viral protein [21] . To investigate whether the STAT1- and STAT3-related activities of IE1 are linked , we subjected our wild-type and mutant IE1 expressing cell lines to immunoblotting for Y701- and S727-phosphorylated STAT1 . Again , all but the IE1dl410-420 and dl421-445 mutants induced STAT1 Y701 and S727 phosphorylation in a fashion comparable to the wild-type viral protein ( Fig 5A ) . Cells expressing these two mutants also exhibited lower overall STAT1 levels compared to cells expressing the wild-type or other mutant proteins , most likely because pSTAT1 positively regulates its own expression . Likewise , when tested for induction of genes representative of pSTAT1-responsive type II ISGs ( CXCL10 and CXCL11 ) by RT-qPCR , the IE1dl410-420 and dl421-445 proteins were severely defective . All other mutants were active for CXCL10 and CXCL11 induction , although IE1dl373-386 and dl387-394 displayed reduced and IE1dl476-491 protein increased activities compared to the wild-type ( Fig 5B ) . We suspect that mutations between amino acids 373 and 394 may indirectly affect IE1-STAT3 complex formation , perhaps by impacting the viral protein’s overall structure . Conversely , increased nucleoplasmic localization may explain why the CTD-deficient IE1 mutant ( IE1dl410-420 ) is more potent than the chromatin-associated viral protein . We also observed that the IE1-dependent repression of genes responsive to STAT3 , IL6 or/and OSM tends to precede the activation of genes responsive to STAT1 or/and IFNγ by IE1 ( S3 Fig ) . These results suggest that the effects IE1 exerts on the STAT1- and STAT3-dependent signaling pathways are related and indicate that the former might depend on the latter . In the next step , we addressed the mechanism underlying the proposed link between the IE1-related effects on STAT3- and STAT1-dependent signaling . To this end , we used two siRNA duplexes each to individually knock-down expression from three essential genes ( STAT3 , IL6ST and JAK1 ) of the IL6-type pSTAT3 signaling pathway and examined the consequences on IE1-mediated induction of pSTAT1-responsive type II ISGs ( CXCL10 and GBP4 ) . Either of the target-specific siRNAs reduced the levels of the corresponding mRNA by ≥80% , compared to a non-specific siRNA , without significantly affecting the IE1 mRNA levels ( Fig 6A–6C ) . Notably , STAT3 knock-down by either of the two specific siRNAs lead to a dramatic increase in CXCL10 and GBP4 transcript levels ( in the absence of IE1 ) and enhanced IE1-dependent ISG induction . Conversely , IL6ST or JAK1 knock-down had little effect on CXCL10 and GBP4 transcript levels , but markedly reduced ISG induction by IE1 ( Fig 6A–6C ) . In comparison , IFNGR1 knock-down affected ISG ( CXCL9 , CXCL10 and CXCL11 ) activation by IE1 only slightly ( S4 Fig ) confirming that this effect is largely independent from upstream components of the type II IFN signaling pathway including IFNγ [21] . These results indicate that the pSTAT1-dependent IFNγ-like transcriptional response observed in the presence of IE1 depends on upstream components of the IL6-type ( but not IFNγ-type ) signaling pathway and may involve targeting of STAT3 . Based on the above results , we hypothesized that IE1 may redirect IL6-related signaling away from STAT3 and towards STAT1 activation . In accordance with this hypothesis , IL6 triggered robust and prolonged STAT1 Y701 phosphorylation in the presence of wild-type IE1 , but not in the presence of IE1dl410-420 or in the absence of the viral protein ( Fig 7A and 7B ) . In fact , the combination of IL6 and IE1 was at least equally efficient in mediating STAT1 activation as IFNγ . Consistently , IL6 also caused ~1 , 000-fold increased expression of pSTAT1-responsive type II ISGs ( CXCL10 and CXCL11 ) in cells with wild-type IE1 compared to cells with IE1dl410-420 or no viral protein ( Fig 7C ) . We further observed that IFNγ was more effective in inducing STAT1 Y701 phosphorylation and type II ISG induction when wild-type IE1 was expressed . Instead , wild-type IE1 inhibited induction of type I ISGs by IFNα in accordance with previous reports [52–54] ( Fig 7B and 7C ) . Finally , wild-type IE1 but not IE1dl410-420 inhibited both basal and cytokine-induced expression of IL6/STAT3-responsive genes ( CXCL12 and SOCS3 ) ( Fig 7C ) . From these results we conclude that IE1 disconnects upstream IL6-type from downstream STAT3-dependent signaling , reconnecting it to downstream STAT1 signaling and related gene expression . To be able to test whether the IE1-STAT3 interaction diverts IL6 signaling also in the context of infection , we derived a mutant virus ( TBIE1dl410-420 ) specifically lacking the sequence encoding IE1 amino acids 410 to 420 from a bacterial artificial chromosome ( BAC ) clone of the hCMV TB40/E strain . In addition , a virus in which the mutated sequence was reverted to wild-type ( TBrvIE1dl410-420 ) was constructed . The integrity and identity of the BACs underlying the mutant and revertant viruses were verified in comparison to the wild-type ( TBwt ) by restriction fragment and PCR analysis , respectively ( S5 Fig ) . Following infection of MRC-5 cells , an increase in nuclear localization of STAT3 compared to mock-infected cells was observed with the TBwt and TBrvIE1dl410-420 but not the TBIE1dl410-420 virus ( S6 Fig ) . At low input multiplicity , the TBIE1dl410-420 virus exhibited attenuated replication and increased sensitivity to exogenous IFNβ compared to the TBwt and TBrvIE1dl410-420 viruses ( Fig 8A ) . At high multiplicity of infection , the IE1dl410-420 protein accumulated to similar levels as the full-length protein expressed from TBwt or TBrvIE1dl410-420 ( Fig 8B ) . However , significant IL6-dependent STAT1 Y701 phosphorylation was only observed with the wild-type and revertant but not the mutant virus at both early ( 24 h ) and late times ( 72 h ) post infection ( Fig 8B and S7A Fig ) . Upon TBwt infection , several type II ISGs ( CXCL9 , CXCL10 , CXCL11 and IDO ) exhibited varying degrees of activation at 24 h post infection in accordance with previous reports [21 , 67–76] . Following addition of IL6 to the infected cells , expression of the four tested ISGs increased by ~2- to >30-fold . This increase was significantly stronger in TBwt and TBrvIE1dl410-420 compared to TBIE1dl410-420 infections ( Fig 8C and S7B Fig ) . These results indicate that the IE1-STAT3 interaction is not only sufficient to rewire IL6-type to IFNγ-like signaling , but also largely required to mediate this response during hCMV infection . The phenotype of the TBdlIE1410-420 mutant virus ( Fig 8A ) suggests that the diverted signaling linked to IE1-STAT3 interaction may be beneficial to hCMV replication . However , STAT3 knock-down turned out to inhibit replication of both TBwt and TBdlIE1410-420 ( S8 Fig ) . Moreover , STAT2 binding maps to a region overlapping IE1 amino acids 410–420 [53] and TBdlIE1410-420 is hypersensitive to IFNβ ( Fig 8A ) . Thus , the replication defect of TBdlIE1410-420 may derive from either or both inhibition of type I IFN signaling ( via STAT2 binding ) or rewiring of IL6/IFNγ signaling ( via STAT3 binding ) . To discriminate between the consequences of STAT2 and STAT3 targeting by IE1 , infections with EGFP expressing hCMV strains were performed in human fibroblasts lacking any STAT2 protein . In the absence of STAT2 , replication of the gTBIE1dl410-420 mutant did not differ from the wild-type ( gTBwt ) and revertant ( gTBrvIE1dl410-420 ) virus . Furthermore , addition of IL6 selectively reduced the replication efficiency of gTBwt and gTBrvIE1dl410-420 but not gTBIE1dl410-420 ( Fig 8D and 8E ) . These results indicate that the type II IFN-like response mediated by IL6-type cytokines in the presence of IE1 moderates rather than promotes hCMV replication , revealing an unanticipated temperance activity encoded in the viral protein .
Decades of research have established that hCMV IE1 is a promiscuous transcriptional activator of viral and cellular genes [42 , 65] . In contrast , evidence supporting transcriptional repression by IE1 is much scarcer . Our study reveals that IE1 is as much a repressor as it is an activator of host gene expression . This finding is unexpected and challenges the view that the viral protein is essentially a transcriptional activator . Many of the genes found to be down-regulated by IE1 have been previously shown to be up-regulated by pSTAT3 or/and upstream regulators in this pathway including cytokines of the IL6 family . Beyond that , we identified several new pSTAT3 targets among the genes that are subject to IE1 repression . We predict that many more of the genes repressed by IE1 than those listed in Table 1 and marked as STAT3-responsive in Table 2 will turn out to be activated by signaling via pSTAT3 . Although , to our knowledge , this is the first systematic genome-wide analysis of transcriptional repression by IE1 , our results are consistent with earlier findings linking the viral protein to down-regulation of the genes encoding glial fibrillary acidic protein [77 , 78] and SOCS3 [30] , both of which are activated by pSTAT3 [79–82] . Our results are also in line with previous reports on inhibition of IL6 signaling by hCMV , although hCMV-dependent activation of IL6 signaling has also been observed [30 , 83] . The possibility that IL6 signaling is induced and subsequently blocked in cells hit by infectious virus ( expressing IE1 ) , but induced without being blocked in bystanding cells hit by non-infectious virus ( not expressing IE1 ) may reconcile these seemingly conflicting findings . Our conclusions regarding the mechanism underlying JAK-STAT-related transcriptional repression and activation by IE1 result partly from the analysis of mutant proteins . These analyses map all activities linked to repression of pSTAT3- and activation of pSTAT1-responsive genes , including IE1-STAT3 complex formation , to a region between amino acids 410 and 445 in the C-terminal domain of the viral protein . This region contains residues upstream of and including AD2 , a low complexity motif rich in glutamic and aspartic acid . Other interactions mapped to the C-terminal domain of IE1 include binding to chromatin/nucleosomes via the CTD ( amino acids 476–491 ) [59 , 61 , 62] , conjugation to SUMO1 via lysine 450 [57 , 58 , 84] and binding to STAT2 [52–54] . The CTD proved to be necessary for SUMO1 conjugation , confirming results from previous reports [85 , 86] and suggesting that nucleosome binding may be a prerequisite for SUMOylation of IE1 . However , nucleosome interaction and SUMOylation were not required for the observed effects IE1 exerts on JAK-STAT signaling . In fact , SUMOylation has been previously shown to negatively regulate the IE1-STAT2 interaction [54] . The relationship between the IE1-STAT2 and IE1-STAT3 interactions is discussed below . Most previous work has suggested that IE1 activates gene expression through various chromatin-directed mechanisms including recruitment of transcription factors [87–89] , inhibition of transcription repressors [90–92] , acetylation of histones [93] or reorganization of nucleosomes [94] . Our present study demonstrates that IE1 also passively modulates transcription upstream of chromatin by linking repression of STAT3- to activation of STAT1-responsive human genes . Our results suggest a model ( Fig 9 ) where STAT3 , which shuttles through the nucleus independent of phosphorylation [5] , forms a nuclear resident complex with IE1 . The nuclear retention imposed on STAT3 precludes the protein from being Y705-phosphorylated by its cytoplasmic kinase JAK1 . Thus , the STAT3 accumulating in the nucleus in the presence of IE1 is mostly unphosphorylated [30] and therefore unable to bind to and activate pSTAT3-dependent target genes . It remains to be investigated whether IE1 also prevents DNA binding of Y705-phosphorylated STAT3 . In the context of activated IL6-type signaling , the nuclear IE1-STAT3 interaction becomes evident as transcriptional repression of pSTAT3-responsive genes . In addition to mediating this repression , the IE1-STAT3 interaction also triggers transcriptional activation . We propose that , in the absence of cytoplasmic STAT3 , JAK1 mediates Y701 phosphorylation of STAT1 leading to nuclear accumulation of pSTAT1 and activation of type II ISGs . Type II ISG activation by IE1 has been observed before [21] , but seems to contradict a recent report showing that the viral protein disrupts type II IFN signaling by an unknown mechanism [95] . It appears that this mechanism involves indirect consequences of IE1-STAT2 rather than direct IE1-STAT1 binding , just as the activation of type II ISGs is an indirect consequence of the IE1-STAT3 interaction . In fact , direct binding between IE1 and STAT1 has not been demonstrated , although the two proteins may indirectly associate via STAT heterodimer formation [52 , 63] . In any case , our model is fully consistent with the fact that depletion of STAT3 phenocopies the IE1-related effects we observe , and that these effects are largely dependent on pSTAT1 [21] , IL6ST and JAK1 but not IFNGR1 . It is also in line with the observation that IE1 promotes STAT1 phosphorylation and nuclear accumulation [21] . Finally , the model fits the idea that IFNγ‐like responses depend on pSTAT1 , but not necessarily on signaling through the cognate receptor . In fact , activation of STAT1 through foreign chimeric receptors proved to be almost as effective in mediating major aspects of an IFNγ response in human cells as activation through the natural receptor [96] . It has also been shown that phosphotyrosine motifs in both IFNGR and IL6ST can serve as docking sites for the STAT1 SH2 domain [97–100] . Two of the four phosphotyrosine motifs in IL6ST are specific for STAT3 while two can recruit both STAT3 and STAT1 with similar affinities [100] . The absence of STAT3 may release competition for the common docking sites , favoring recruitment and activation of STAT1 . In accordance with this idea , IL6 triggered an IFNγ-like response including prolonged activation of STAT1 and induction of multiple type II ISGs in mouse cells lacking STAT3 [15] . A similar type II ISG response to STAT3 depletion was observed in human cells ( this work ) . The STAT3/STAT1 cross-regulation seems to result in an integrated signal that can be fine-tuned depending on the cellular context , strongly arguing for flexible rather than fixed wiring of these pathways [14 , 16 , 17 , 101–103] . We anticipate that future work will recognize at least a subset of genes repressed or activated by IE1 as important factors in hCMV infection . In fact , several targets of IE1 repression identified in this study have already been ascribed critical roles in the hCMV life cycle . To give but one example , IFNγ-inducible protein 16 ( IFI16 ) acts as a foreign DNA sensor and restriction factor for hCMV [104 , 105] , suggesting that IE1 might promote viral replication by limiting its expression . In accordance with this speculation , replication of an IE1 mutant hCMV ( TBIE1dl410-420 ) deficient for STAT3/STAT1 pathway diversion is attenuated compared to wild-type and revertant viruses . It should be noted , however , that the attenuated phenotype of the TBIE1dl410-420 virus does not necessarily result from repression of STAT3-responsive genes . In fact , this study mapped STAT3 binding between amino acids 410 and 445 , while our previous work mapped STAT2 binding between amino acids 373 and 445 in IE1 [51] . Thus , STAT3 and STAT2 seem to bind to the same or closely adjoining sites in IE1 , making it difficult to work out the extent by which either interaction contributes to the observed replication defect . That said , the replication defect exhibited by TBIE1dl410-420 in normal fibroblasts was entirely rescued in STAT2-deficient cells . Moreover , TBIE1dl410-420 proved to be more sensitive to exogenous IFNβ compared to wild-type and revertant viruses in normal fibroblasts . These observations indicate that the attenuation of TBIE1dl410-420 results from the lack of IE1-STAT2 interaction , which would normally promote viral replication by inhibiting type I IFN signaling . In contrast , the type II IFN-like response linked to IE1-STAT3 interaction appears to moderate rather than promote hCMV replication consistent with reports that IFNs govern viral latency , at least in murine cytomegalovirus ( mCMV ) [106 , 107] . We are therefore tempted to speculate that , by imparting a low-level chronic IFNγ-like response , IE1 might contribute to viral quiescence in latently infected cells exposed to cytokines or growth factors that activate IL6ST . This idea adds to the emerging concept that IE1 , a protein traditionally linked exclusively to the viral productive cycle , may also have an important role in hCMV latency [41 , 101] . IL6 and IFNγ are among the most pivotal cytokines in shaping innate and adaptive host responses to infectious pathogens [7 , 13] . In many instances , the two cytokines and their pathways have been associated with the outcomes of hCMV or mCMV infection and pathogenesis [30 , 108–113] . In addition to IL6-type cytokines , many other important factors signal via the IL6ST-JAK1-STAT3 axis . These factors include GCSF and IL10 , both of which have been linked to hCMV infection including viral latency and reactivation [114–116] . Signaling through STAT3 or STAT1 usually results in distinct and sometimes opposing outcomes . For instance , STAT1 tends to promote apoptosis in a variety of cell types whereas STAT3 typically has anti-apoptotic effects . Likewise , STAT1 usually acts anti-proliferative while STAT3 rather promotes cellular proliferation [14 , 17] . By merging two major signaling pathways with diverse actions in many cell and tissue types , IE1 may impact hCMV infection and pathogenesis in surprising ways that future work will explore .
The pMD2 . G and psPAX2 packaging vectors for lentivirus production were obtained from Addgene ( plasmids #12259 and #12260 , respectively ) . The lentiviral plasmid pLKOneo . CMV . EGFPnlsTetR , encoding the tetracycline repressor linked to a nuclear localization signal and the enhanced green fluorescent protein ( EGFP ) [117] , was kindly provided by Roger Everett ( University of Glasgow ) . Plasmids pCMV . TetO . cIE1 and pLKO . DCMV . TetO . cIE1 ( herein referred to as pLKO . DCMV . TetO . IE1 ) , expressing hCMV ( Towne ) IE1 under the control of a tetracycline- or dox-inducible promoter , have been described [21] . Variants of these plasmids encoding wild-type IE1 linked to an N-terminal hemagglutinin ( HA ) epitope tag ( pLKO . DCMV . TetO . HA-IE1 ) or HA-tagged IE1 deletion mutants were constructed by standard PCR ( IE1 , IE1dl476-491 , IE19 , NLS-IE1dl1-404 , 2×Stop-IE1 ) or overlap extension PCR ( IE1dl373-386 , IE1dl387-394 , IE1dl395-409 , IE1dl410-420 , IE1dl421-445 , IE1dl446-450 and IE1dl451-475 ) using oligonucleotide primers listed in Table 3 . For each new construct , the entire IE1-specific nucleotide sequence was verified by DNA sequencing . The lentiviral plasmids used for inducible expression of human STAT3- or firefly luciferase-specific shRNAs ( shSTAT3_1 , shSTAT3_2 and shLUC ) were generated by ligating annealed oligonucleotides ( Table 3 ) to EcoRI- and AgeI-digested Tet-pLKO-puro ( Addgene plasmid #21915 ) , and the resulting constructs were verified by DNA sequencing . First strand cDNA of the full-length human STAT3α coding sequence was prepared from total RNA extracted from MRC-5 cells using an oligo ( dT ) 20 primer . The STAT3α cDNA was linked to a C-terminal Myc epitope tag and PCR-amplified using oligonucleotide primers #961 and #962 ( Table 3 ) . The purified PCR product was ligated to HpaI-digested , dephosphorylated retroviral vector pLHCX ( Clontech , 631511 ) resulting in plasmid pLHCX-STAT3α-Myc . QuikChange site-directed mutagenesis of pLHCX-STAT3α-Myc using oligonucleotide primers #1051 and #1052 ( Table 3 ) resulted in plasmid pLHCX-STAT3α_Y705F-Myc encoding a trans-dominant negative STAT3α-Myc variant resistant to phosphorylation at Y705 [66] . For both new constructs , the correct orientation and nucleotide sequence of the STAT3 insert were verified by DNA sequencing . To generate expression plasmids for proteins linked to N-terminal mCherry , the mCherry cDNA was PCR-amplified from plasmid pIM-Asf1B-mCherry ( kindly provided by Jean-Yves Thuret , Paris-Sud University ) using oligonucleotide primers #1201 and #1202 ( Table 3 ) and ligated to BglII- and HindIII-digested pCMV . TetO . HA-IE19 . The HA-IE19 insert of the resulting construct was released with HindIII and EcoRI and replaced with the HA-IE1 coding sequences of plasmids pCMV . TetO . HA-IE1 , pCMV . TetO . HA-2×Stop-IE1 and pCMV . TetO . HA-NLS-IE1dl1-404 . The resulting constructs were verified by DNA sequencing . Plasmid templates pLAY2 and pUC-MIE-Kan_I-SceI for generation of mutant and revertant hCMV TB40/E BACs by en passant mutagenesis , kindly provided by Karsten Tischer ( Freie Universität Berlin ) , have been described [62 , 94 , 118] . The primary antibodies used in this work were as follows: mouse anti-HA ( Covance , MMS-101P ) , mouse anti-IE1 ( [119] , 1B12 ) , mouse anti-IE1/IE2 ( Millipore , MAB810R ) , mouse anti-Myc ( Santa Cruz Biotechnology , sc-40 ) , rabbit anti-glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) ( Abcam , ab9485 ) , rabbit anti-H2B ( Abcam , ab1790 ) , rabbit anti-STAT1 p84/p91 ( Santa Cruz Biotechnology , sc-346 ) , rabbit anti-STAT1α p91 ( Santa Cruz Biotechnology , sc-345 ) , rabbit anti-STAT2 ( Santa Cruz Biotechnology , sc-22816 ) , rabbit anti-STAT3 ( Santa Cruz Biotechnology , sc-482x ) , rabbit anti-STAT3α ( Cell Signaling Technologies , 8768 ) , rabbit anti-pSTAT1 ( Y701 ) ( Cell Signaling Technologies , 9171 ) , rabbit anti-pSTAT1 ( S727 ) ( Cell Signaling Technologies , 9177 ) , rabbit anti-pSTAT3 ( Cell Signaling Technologies , 9145 ) , rabbit anti-SUMO1 ( Santa Cruz Biotechnology , sc-9060 ) and mouse anti-α-tubulin ( TUBA ) ( Thermo Fisher Scientific , A-11126 ) . The following secondary antibodies were used: peroxidase-conjugated goat anti-mouse immunoglobulin G ( IgG ) ( Dianova , 115-035-166 ) or goat anti-rabbit IgG ( Dianova , 111-035-144 ) for immunoblotting , and highly cross-adsorbed Alexa Fluor 488-conjugated goat anti-mouse IgG ( Thermo Fisher Scientific , A-11001 ) or Alexa Fluor 594-conjugated goat anti-rabbit IgG ( Thermo Fisher Scientific , A-11037 ) for immunofluorescence . MRC-5 human embryonic lung fibroblasts ( American Type Culture Collection , CCL-171 ) , STAT2-deficient primary skin fibroblasts from a 5-year-old child with a history of disseminated vaccine-strain measles [120] ( kindly provided by Sophie Hambleton , Newcastle University ) and Phoenix-Ampho retrovirus packaging cells ( kindly provided by Garry Nolan , Stanford University ) were maintained in Dulbecco's Modified Eagle’s Medium ( DMEM ) supplemented with 10% fetal calf serum ( FCS ) , 100 units/ml penicillin and 100 μg/ml streptomycin . The human embryonic kidney cell line 293T ( GenHunter , Q401 ) was cultured in the same medium containing 400 μg/ml G418 sulfate . All cultures were regularly screened for Mycoplasma sp . using an in-house qPCR assay . Where applicable , cells were treated with 1 , 000 U/ml recombinant human IFNα A/D ( R&D Systems , 11200 ) , 1 , 000 U/ml IFNβ1α ( Miltenyi Biotec , 130-107-888 ) , 1 , 000 U/ml recombinant human IFNγ ( R&D Systems , 285-IF ) , 100 ng/ml recombinant human OSM ( R&D Systems , 295-OM-010 ) or 100 ng/ml recombinant human IL6 ( R&D Systems , 206-IL-010 ) plus 100 ng/ml recombinant human IL6Rα ( R&D Systems , 227-SR-025 ) for the indicated durations . Fibroblasts are normally unresponsive to IL6 , since they have little or no IL6R ( also known as IL6Rα ) . However , soluble IL6R can bind IL6 with similar affinity as membrane-bound IL6R and the complex of IL6 and soluble IL6R can interact with and signal through IL6ST ( also known as IL6Rβ or gp130 ) . Production of replication-deficient retroviral particles , retrovirus infections and selection of stable cell lines were performed as previously described [21] with minor modifications . Lentiviral particles were generated by transient transfection of 293T cells using calcium phosphate co-precipitation [121] . Recombinant viruses were collected 48 h after transfection and used for two consecutive transductions of 4 h each . To generate TetR cells , low-passage MRC-5 cells were transduced with pLKOneo . CMV . EGFPnlsTetR-derived lentiviruses and selected with G418 sulfate ( 0 . 3 mg/ml ) . To generate TetR-IE1 cells , TetR cells were transduced with pLKO . DCMV . TetO . HA-IE1-derived lentiviruses and selected with puromycin ( 1 μg/ml ) . TetR cells were maintained in medium containing G418 sulfate ( 0 . 3 mg/ml ) , while TetR-IE1 cells were cultured in the presence of both G418 sulfate ( 0 . 3 mg/ml ) and puromycin ( 1 μg/ml ) . To induce IE1 expression , cells were treated with dox ( Clontech , 631311 ) at a final concentration of 1 μg/ml . To generate cells with inducible expression of human STAT3- or firefly luciferase-specific shRNAs , MRC-5 cells were transduced with Tet-pLKO-puro . shSTAT3_1- , Tet-pLKO-puro . shSTAT3_2- or Tet-pLKO-puro . shLUC-derived lentiviruses and selected in medium containing puromycin ( 1 μg/ml ) . To induce shRNA expression , cells were treated with 1 μg/ml dox . At least half of the culture medium was replaced every 48 h with fresh dox added to maintain stable shRNA expression . To generate cells constitutively expressing STAT3α-Myc proteins , TetR cells were infected with pLHCX-derived retroviruses encoding STAT3α-Myc or STAT3α_Y705F-Myc and selected in medium containing hygromycin B ( 0 . 2 mg/ml ) . Wild-type virus of the low passage hCMV strain TB40/E ( TBwt ) was derived from TB40-BAC4 [122] , kindly provided by Christian Sinzger ( Ulm University ) . A modified version of this BAC with an SV40-EGFP-BGH PolyA cassette inserted between the US34 and TRS1 genes [123] was kindly provided by Felicia Goodrum ( University of Arizona ) and used to generate the EGFP expressing TB40/E wild-type virus gTBwt . Mutant BACs encoding IE1 with internal deletion of amino acids 410 to 420 ( pTBIE1dl410-420 and pgTBIE1dl410-420 ) and corresponding revertant BACs ( pTBrvIE1dl410-420 and pgTBrvIE1dl410-420 ) were generated from pTBwt or pgTBwt by markerless en passant mutagenesis , as previously described [62 , 94] using plasmids pLAY2 and pUC-MIE-Kan_I-SceI and oligonucleotide primers listed in Table 3 . The identity and integrity of each BAC were verified by DNA sequencing of the modified region and restriction fragment length analysis following digestion with EcoRI , respectively . Viruses were reconstituted and virus stocks produced upon electroporation of MRC-5 cells with BAC DNA following standard protocols . Titers were determined by plaque assay , and quantification of intracellular viral genome equivalents was performed as described [53] except that cells were cultured in the presence of phosphonoacetic acid ( 0 . 2 mg/ml ) for 18 h before DNA isolation and qPCR analysis . MRC-5 lung and STAT2-deficient skin fibroblasts were grown to confluency , serum-starved in DMEM containing 0 . 5% FBS for 24 h and infected in medium with 0 . 5% FBS . After 6 h the inoculum was removed , cells were rinsed with pre-warmed DMEM and cultured in serum-reduced ( 0 . 5% FBS ) DMEM . Silencer Select siRNAs ( chemically modified , 21-mer , locked nucleic acid , double-stranded RNAs; Thermo Fisher Scientific ) at a final concentration of 30 nM were introduced into cells using the Lipofectamine RNAiMAX Reagent ( Thermo Fisher Scientific ) and following the manufacturer’s protocols . Briefly , exponentially growing cells were seeded either in 12-well dishes at 2 . 5×105 cells/well for RNA analyses or in 6-well dishes at 5×105 cells/well for protein analyses . Transfections were performed in Opti-MEM I Reduced Serum Medium ( Thermo Fisher Scientific ) with 2 μl or 5 μl transfection reagent for 12 or 6 wells , respectively . The siRNA sequences are listed in Table 4 . The transcriptome analysis of TetR and TetR-IE1 cells using Affymetrix Human Gene 1 . 0 ST Arrays has been described [21] , and the complete set of data is accessible through Gene Expression Omnibus ( National Center for Biotechnology Information ) , Series GSE24434 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE24434 ) . To determine steady-state mRNA levels by RT-qPCR , total RNA was isolated from fibroblasts cultured in 12-well dishes using the RNeasy Mini Kit and RNase-Free DNase Set ( Qiagen ) according to the manufacturer’s protocols . First-strand cDNA was synthesized at 50°C using the AffinityScript Multiple Temperature cDNA Synthesis Kit and oligo ( dT ) primers ( Agilent Technologies ) starting with equal amounts of total RNA . First-strand cDNA was diluted 10-fold in sterile ultrapure water , and 5 μl were used for real-time PCR exactly as described in detail in previous publications [21 , 30] . The sequences of oligonucleotide primers used for qPCR are listed in Table 3 . Preparation of whole cell protein extracts , sodium dodecyl sulfate ( SDS ) -polyacrylamide gel electrophoresis and immunoblotting were performed according to previously published protocols [53 , 93] . Subcellular fractionation was done as described [21] . For indirect immunofluorescence microscopy , cells were seeded on high-precision cover glasses with thickness No . 1 . 5H ( VWR , MARI 0117640 ) and processed as described [21] . Following immunostaining , samples were covered with ProLong Gold Antifade Mountant with DAPI ( Molecular Probes , P36931 ) , and deconvoluted images were acquired using a BZ-9000 Biorevo all-in-one fluorescence microscope ( Keyence ) . For SUMOylation analysis , about 7×106 cells were lysed on ice in 500 μl buffer containing 50 mM Tris-HCl pH 7 . 5 , 150 mM NaCl , 2 mM MgCl2 , 0 . 5% ( v/v ) Igepal CA-630 , 1% ( v/v ) Triton X-100 , 1% ( v/v ) protease inhibitor cocktail III ( Calbiochem ) , 1% ( v/v ) phosphatase inhibitor cocktail II ( Calbiochem ) and 5 μg/ml N-ethylmaleimide . After sonification in a Branson Sonifier 450 cup horn resonator ( 25 pulses at duty cycle 80% and output control 10 ) , lysates were cleared by centrifugation ( 20 , 000×g , 25 min , 4°C ) and incubation with 20 μl mouse IgG agarose ( Sigma-Aldrich ) . Soluble material was reacted with 10 μl anti-HA agarose ( Sigma-Aldrich ) for 2 h at 4°C with gentle rotation . Immune complexes were washed three times in modified radioimmunoprecipitation ( RIPA ) buffer ( 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 0 . 1% [w/v] SDS , 1% [v/v] Igepal CA-630 , 0 . 5% [w/v] sodium deoxycholate ) and once in nuclease reaction buffer ( 50 mM Tris-HCl pH 8 . 0 , 2 mM MgCl2 ) before incubation with 25 U Benzonase nuclease ( Novagen ) in a volume of 100 μl for 30 min . After two additional washing steps in RIPA buffer , proteins were eluted by addition of 80 μl 1× sample buffer [124] and incubation for 10 min at 95°C . For co-immunoprecipitation analyses involving dox-treated TetR-IE1 fibroblasts , resting cells of two 15-cm dishes were cross-linked by treatment with 1% ( v/v ) formaldehyde for 10 min at room temperature . Monolayers were washed twice in ice-cold serum-free DMEM , and cells were scraped in DMEM supplemented with 1% protease inhibitor cocktail III , pelleted by centrifugation ( 1 , 200×g , 10 min , 4°C ) and lysed in 1 . 5 ml RIPA buffer supplemented with 1% protease inhibitor cocktail III . After sonification ( 30 pulses at duty cycle 80% and output control 10 in a Branson Sonifier 450 cup horn resonator ) , lysates were cleared by centrifugation ( 20 , 000×g , 30 min , 4°C ) . Supernatants were incubated with 150 μl Protein A Agarose/Salmon Sperm DNA slurry ( Millipore ) for 45 min at 4°C . After centrifugation ( 14 , 000×g , 10 min , 4°C ) , 700 μl supernatant were incubated with 10 μg IE1/IE2 8B1 . 2 antibody or normal mouse IgG for 16 h at 4°C with gentle rotation . Antigen-antibody complexes were isolated by addition of 60 μl Protein A Agarose/Salmon Sperm DNA slurry . Complexes were washed once in RIPA buffer , and nucleic acids were removed by incubation with 25 U Benzonase nuclease in 100 μl nuclease reaction buffer for 30 min at 4°C . After five additional washing steps in RIPA buffer , bound proteins were eluted by incubation in 45 μl 1× sample buffer for 10 min at 99°C . For co-immunoprecipitations involving plasmid-transfected 293T cell cultures , cells of a 10-cm dish were harvested in PBS without prior formaldehyde cross-linking . Cells were pelleted by centrifugation ( 550×g , 8 min , 4°C ) and resuspended in 0 . 5 ml CoIP lysis buffer ( 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 10% ( v/v ) glycerol , 0 . 5% ( v/v ) Triton X-100 ) supplemented with 1% protease inhibitor cocktail III . Lysates were incubated on ice for 10 min and cleared by centrifugation ( 20 , 000×g , 30 min , 4°C ) . Supernatants were incubated with 25 μl Pierce Anti-HA Magnetic Beads ( ThermoFisher Scientific , 88836 ) for 2 h at 4°C with gentle rotation . Complexes were washed five times in CoIP wash buffer ( 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 0 . 1% ( v/v ) Igepal CA-630 ) and eluted by incubation in 50 μl 1× sample buffer for 5 min at 95°C . ChIP coupled to qPCR was performed essentially as described [21 , 30] with minor modifications . Briefly , cells were cross-linked for 15 min at 37°C . Sheared chromatin was centrifuged for 30 min to remove insoluble material , and the supernatant from 7×106 cells was subjected to a pre-clearing step with 75 μl Protein A Agarose/Salmon Sperm DNA slurry for 30 min at 4°C with gentle rotation . Immunoprecipitations were carried out using 10 μg STAT3 C-20 antibody or normal rabbit IgG . Oligonucleotide primers used for subsequent qPCRs are listed in Table 3 . The Core Analysis function ( default analysis settings ) of the Ingenuity Pathway Analysis software application ( Content version 24718999 , Build version 366632M; Qiagen ) was used to explore upstream regulators in the human transcriptome repressed by IE1 relative to an Affymetrix Human Gene 1 . 0 ST Array reference set . Only direct and indirect relationships where ‘confidence = Experimentally Observed’ were considered . | Our previous work has shown that the human cytomegalovirus ( hCMV ) major immediate-early 1 protein ( IE1 ) modulates host cell signaling pathways involving proteins of the signal transducer and activator of transcription ( STAT ) family . IE1 has also long been known to facilitate viral replication by activating transcription . In this report we demonstrate that IE1 is as significant a repressor as it is an activator of host gene expression . Many genes repressed by IE1 are normally induced via STAT3 signaling triggered by interleukin 6 ( IL6 ) or related cytokines , whereas many genes activated by IE1 are normally induced via STAT1 signaling triggered by interferon gamma ( IFNγ ) . Our results suggest that the repression of STAT3- and the activation of STAT1-responsive genes by IE1 are coupled . By targeting STAT3 , IE1 rewires upstream STAT3 to downstream STAT1 signaling . Consequently , genes normally induced by IL6 are repressed while genes normally induced by IFNγ become responsive to IL6 in the presence of IE1 . We also demonstrate that , by switching an IL6 to an IFNγ-like response , IE1 tempers viral replication . These results suggest an unanticipated dual role for IE1 in either promoting or limiting hCMV propagation and demonstrate how a key viral regulatory protein merges two central cellular signaling pathways to divert cytokine responses relevant to hCMV pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"phosphorylation",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"immunology",
"microbiology",
"viruses",
"developmental",
"biology",
"dna",
"viruses",
"molecular",
"development",
"molecular",
"biology",
"techniques",
"herpesviruses",
"research",
"and",
"analysis",
"methods",
"human",
"cytomegalovirus",
"small",
"interfering",
"rnas",
"artificial",
"gene",
"amplification",
"and",
"extension",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"stat",
"signaling",
"viral",
"replication",
"molecular",
"biology",
"immune",
"system",
"biochemistry",
"signal",
"transduction",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"post-translational",
"modification",
"polymerase",
"chain",
"reaction",
"virology",
"viral",
"pathogens",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"cell",
"signaling",
"organisms"
] | 2016 | Human Cytomegalovirus Immediate-Early 1 Protein Rewires Upstream STAT3 to Downstream STAT1 Signaling Switching an IL6-Type to an IFNγ-Like Response |
The cooperative developmental system of the social amoeba Dictyostelium discoideum is susceptible to exploitation by cheaters—strains that make more than their fair share of spores in chimerae . Laboratory screens in Dictyostelium have shown that the genetic potential for facultative cheating is high , and field surveys have shown that cheaters are abundant in nature , but the cheating mechanisms are largely unknown . Here we describe cheater C ( chtC ) , a strong facultative cheater mutant that cheats by affecting prestalk differentiation . The chtC gene is developmentally regulated and its mRNA becomes stalk-enriched at the end of development . chtC mutants are defective in maintaining the prestalk cell fate as some of their prestalk cells transdifferentiate into prespore cells , but that defect does not affect gross developmental morphology or sporulation efficiency . In chimerae between wild-type and chtC mutant cells , the wild-type cells preferentially give rise to prestalk cells , and the chtC mutants increase their representation in the spore mass . Mixing chtC mutants with other cell-type proportioning mutants revealed that the cheating is directly related to the prestalk-differentiation propensity of the victim . These findings illustrate that a cheater can victimize cooperative strains by exploiting an established developmental pathway .
Cooperative behaviors are susceptible to exploitation by cheaters – individuals that do not pay the full cost of cooperation , but reap the benefits [1] and thus take advantage of other cooperative individuals ( victims ) . Cheating is predicted to affect the relative fitness of any interacting partners , especially when multiple genotypes are involved . Such behavior is thought to occur in all cooperative societies , and has been demonstrated in several different social insect colonies [2] , [3] , where a significant part of the population ( the workers ) does not take part in reproduction . In these systems , cheaters can manipulate developmental processes , thereby changing the balance between the reproductive ( queen ) and supporting ( worker ) castes . For example , cheaters exploit cooperative genotypes by tweaking mechanisms such as the regulation of organism size [3] , developmental timing [2] and differentiation into different castes [3] , [4] . It is likely that the regulation of other developmental processes – cell-division and cell-fate determination , proportioning and maintenance – is also susceptible to cheating . However , it is hard to study these mechanisms at the genetic and cellular levels due to the complex nature of these social systems . Social microorganisms are good model systems for the study of cheating mechanisms at the molecular level . The social amoebae Dictyostelium discoideum provide an added advantage because the cells exhibit social behavior in the context of multicellular development . Dictyostelium cells propagate as unicellular amoebae and feed on bacteria . However , under conditions of starvation , about 105 cells aggregate and go through multicellular development . The cells give rise to a structure called the fruiting body where 70–80% of the cells form viable spores that may germinate in the next generation to form amoebae , while the remaining cells give rise to dead , vacuolated cells that contribute to stalk-formation and hence sacrifice their reproduction [5] . This developmental cycle is different from the development of metazoan organisms , since multicellularity is achieved by aggregation rather than by cell division of a fertilized egg . An important consequence is that Dictyostelids readily form organisms containing multiple clones . In such chimerae , different genotypes can contribute differently to the production of the reproductive ( spores ) and supporting ( stalk cells ) cell-types , and change their representation in subsequent generations , similar to the cheating behavior seen in insect societies . Disproportionate over-representation of a specific genotype in the spore population of a chimeric fruiting body at the cost of another strain is defined as cheating , and the over- and under-represented strains are termed as ‘cheaters’ and ‘victims’ , respectively . Chimerism has been observed in nature [6] , and clones isolated from the wild can cheat on one another in the laboratory [7] . The first cheater mutant identified in D . discoideum , chtA ( fbxA ) , is an obligate parasite that is unable to form spores in clonal populations [8] . When mixed with chtA , the wild-type prespore cells differentiate into stalk cells . This is the only cheating mechanism that has been identified in Dictyostelium to date . However , since chtA does not complete development under clonal conditions , it is unlikely that its behavior is characteristic of cells in the wild since Dictyostelium strains are often found in clonal populations [6] . Recent studies have shown that a large number of mutations in Dictyostelium can lead to facultative cheating [9] . Facultative cheater mutants are capable of forming fruiting bodies in clonal populations , but cheat on wild-type cells in chimera . These mutants probably cheat by exploiting a variety of mechanisms , and the social genes identified are predicted to be involved in a variety of different cellular processes [9] . Development in Dictyostelium involves both the initial differentiation and proportioning of several different cell-types , and the subsequent maintenance of cell-fate and cell-type proportions . Any of these developmental mechanisms might be co-opted by selfish cheater mutants , akin to what is seen in insect societies . Consequently , the study of such cheater mutants is likely to facilitate greater understanding of specific pathways of differentiation in Dictyostelium , in addition to developmental cheating mechanisms in general . We have studied chtC [10] , one of the strongest facultative cheater mutants identified by Santorelli et al . [9] . We found that chtC has defects in maintaining the prestalk cell fate , and consequently is defective in the expression of certain late prestalk markers . Even though this does not lead to any discernible stalk defects when chtC mutants develop on their own , wild-type cells increase their prestalk differentiation in chimerae with chtC and are cheated upon . These findings suggest that cheaters in Dictyostelium can manipulate mechanisms of developmental regulation such as the maintenance of cell-type proportioning to take advantage of other strains in the population , while retaining their fitness under clonal conditions .
LAS5 was one of the strongest cheater strains identified in a large scale screen for cheater mutants [9] . This mutant strain has a plasmid insertion in the chtC gene [10] . The chtC gene is predicted to encode an approximately 75 kDa protein with a signal peptide anchor and a transmembrane domain at the N-terminus ( Figure 1A ) . This protein has orthologs of unknown function with about 20% identity in ciliates such as Paramecium tetraurelia and Tetrahymena thermophila , but no detectable homology to proteins in other organisms ( data not shown ) . The gene is also up-regulated in AX2 cells incubated with E . coli when compared to cells incubated in axenic medium [11] . To determine the expression properties of the chtC transcript , we collected RNA from AX4 cells at 4-hour intervals throughout development and performed quantitative reverse transcription PCR ( Q-RT-PCR ) with chtC-specific primers ( Figure 1B ) . We found chtC mRNA at all times with a peak at 12 hours of development , when the cells were at the tight aggregate stage , followed by a decline at 16 hours and comparatively lower levels thereafter . We also tested the spatial expression pattern of chtC by whole mount in situ RNA hybridization . The chtC mRNA was uniformly abundant in all cells during the finger stage of development ( data not shown ) , but became highly enriched in the stalk , with the highest levels in the funnel ( Figure 1C ) , which is at the top of the stalk tube , during late culmination ( fruiting body formation ) . The original mutant , LAS5 , had a pBSR1 plasmid insertion at nucleotide 1377 of the chtC ORF ( Figure 1A ) . We generated two new alleles of chtC . The chtCins mutant contains a plasmid insertion at position 1377 of the endogenous locus and the chtCdel mutant contains a plasmid instead of the endogenous region that codes for amino acid 13 – 642 ( Figure 1A ) . Both strains were made sensitive to Blasticidin S to facilitate the analysis of chimerae . The alleles were confirmed by Southern blot analysis and by PCR across the relevant insertion sites ( data not shown ) . We first tested the spore-forming ability of the chtC mutants . Sporulation of the clonal chtC mutants and of 1∶1 chimerae between the chtC mutants and AX4 , were indistinguishable from that of clonal AX4 populations , as tested by determining spore morphology ( data not shown ) , sporulation efficiency , resistance to detergent , and germination efficiency ( Figure S1 ) . This finding is in contrast to the original LAS5 mutant which had a higher sporulation efficiency compared to AX4 cells [9] , suggesting that different alleles of chtC can lead to distinct phenotypes . We then studied the behavior of the chtC mutants in chimera . We first tested whether the chtC mutants co-aggregate with wild-type cells by observing 1∶1 mixtures of either chtCins or chtCdel with AX4 at 8 hours of development ( Figure S2 ) . Both the chtC mutants co-aggregated with AX4 cells , similar to an AX4 control . We then tested the cheating behavior of the chtC mutants by mixing either chtCins or chtCdel at a 1∶1 ratio with AX4/[act15]:GFP ( AX4-GFP ) cells and letting the mixed populations complete development to form fruiting bodies . We also mixed AX4-GFP cells with unlabeled AX4 cells as a control . Following development , we collected all the cells , selected for spores by detergent treatment , and counted the ratio of fluorescent to non-fluorescent spores . In the control mixes we found that the AX4 cells form approximately 50% of the spores , suggesting that the AX4-GFP strain behaves in an almost identical fashion to AX4 ( Figure 1D ) . Both the chtCins and the chtCdel mutants cheated - they formed a significantly higher number of spores than AX4 ( Figure 1D ) . Further , the chtCins mutant cheated significantly more than the chtCdel mutant did , suggesting that the chtCins mutant is not a null mutant . We then tested the two mutants by developing them in a 1∶1 mixture with each other . The chtCins mutant cheated on the chtCdel mutant by forming 60 . 7%±5 . 7% spores , which is significantly greater than the hypothesized value of 50% ( n = 3 , one-sample one-sided t-test , P = 0 . 041 ) . Thus the chtCins mutant is distinct from the chtCdel mutant , suggesting that it is not a null , but possibly a gain-of-function allele . In order to test this possibility further , we performed Northern blot analysis with a chtC probe on 8-hour RNA samples from AX4 and from the chtCins and chtCdel mutants ( Figure 1E ) . The wild type chtC transcript size is 2 kb , as expected from the predicted gene model . The chtCdel mutant does not express detectable levels of the transcript , consistent with the deletion of nearly the entire gene and confirming the hypothesis that it is a null-mutant . The chtCins strain expresses a 5–6 kb transcript . Northern blot analysis with a probe against the inserted plasmid showed that this was due to read-through transcription into the plasmid insertion ( data not shown ) . We also performed RT-PCR with primers against the region of the chtC gene downstream of the insertion and observed a product ( data not shown ) . These data suggest that the chtCins mutant expresses an aberrant transcript that extends across the inserted plasmid and back into the chtC gene . The cheating behavior of the mutant strains and the stalk-enriched expression of the chtC mRNA during late developmental stages suggested that chtC may play a role in stalk development although the ubiquitous expression of the gene at earlier stages may imply a role in prespore cells or spores as well . Nevertheless , the chtC mutant strains appear morphologically indistinguishable from the parental AX4 strain during growth and development , ( Figure S1 and data not shown ) . We therefore tested other stalk phenotypes of the chtC mutants . During development of wild-type D discoideum , the small molecule DIF-1 ( Differentiation Inducing Factor-1 ) induces the differentiation of stalk cells , and inhibits spore-differentiation , and sensitivity to this molecule is important for the differentiation of a specific sub-type of prestalk cells . After differentiation , prestalk cells are localized in the anterior part of a developing slug , where they are required for proper slug migration . Finally , wild-type fruiting bodies in D . discoideum contain stalks that consist of vacuolated cells and cellulose deposits , which are important for the formation of a properly structured stalk [5] . We tested each of these stalk phenotypes in the chtC mutants by examining squashes of culminants ( fruiting bodies ) using high-power phase-contrast microscopy , staining for cellulose with the fluorescent dye calcofluor [12] , testing for DIF-1 sensitivity by the cAMP-removal and 8-Br-cAMP monolayer assays [13] , and testing slug migration . We found no significant difference between AX4 and the chtC mutants in these assays ( data not shown ) . To study stalk differentiation in greater detail , we used 2 different prestalk markers – tagB and ecmA . Expression of the tagB gene is induced 8 hours into development in prestalk cells , about 4 hours earlier than ecmA [14] . Also , unlike ecmA , expression of the tagB gene is not induced by DIF-1 [15] . We developed [tagB]:lacZ labeled strains of both the chtCins and chtCdel mutants , and stained for β-galactosidase activity . In AX4 cells , tagB is expressed in the entire prestalk region [14] . Both the chtCins and chtCdel mutants showed strong staining in the posterior half of the prestalk region ( the prestalk-O or PST-O region [16] ) , and weaker staining in the anterior half ( the prestalk-A or PST-A region ) . There was also significant staining in the prespore region , suggesting that some prespore cells express the tagB marker or have expressed it prior to becoming prespore cells ( Figure 2A ) . In order to test this possibility , we examined the spores made by chtC mutants labeled with the [tagB]:lacZ marker . We found a 100-fold increase in the proportion of tagB-positive spores formed by either of the chtC mutants , compared to AX4 ( Table 1 ) . The deficit of [tagB]:lacZ-expressing cells in the PST-A region , combined with the increase in prespore cells that express [tagB]:lacZ suggests that the tagB-expressing prestalk cells , which contribute to the PST-A region in the wild type , are undergoing transdifferentiation and form spores instead of stalk cells . An increase in this transdifferentiation in the presence of AX4 cells would be a potential mechanism of cheating . However , we observed no significant change in the proportion of [tagB]:lacZ-positive spores when the chtC mutants were mixed with unlabeled AX4 instead of the unlabeled chtC mutant cells ( Table 1 ) . Prolonged migration of Dictyostelium slugs results in increased transdifferentiation of prestalk cells into spores [17] , [18] . To test whether the chtC mutants showed increased transdifferentiation under such conditions , we allowed the [tagB]:lacZ labeled chtC mutants to migrate for 48 hours , and then induced culmination . We collected spores , stained them with X-gal , and counted the number of stained spores ( Table 1 ) . In the chtC mutant strains , 8–12% of the spores were labeled , suggesting that they had a prestalk history . Thus , a significant proportion of the chtC mutant population undergoes a cell-fate transformation , suggesting that the chtC gene is required for the maintenance of the prestalk cell fate . In order to further dissect the prestalk properties of the chtC mutants , we generated chtC mutant strains expressing lacZ under the prestalk promoter , ecmA . We developed these strains , and stained for β-galactosidase activity . Both the chtCins and chtCdel mutants showed strong staining in the PST-O region , but weaker staining in the PST-A region ( Figure 2B ) , similar to the phenotype seen in the [tagB]:lacZ strains , suggesting that in the chtC mutants , the cells in the PST-A region are defective in both tagB as well as ecmA expression . However , there was no discernible change in the expression of ecmA in the prespore region , compared to AX4 . We quantified this phenotype by dissociating the structures during late culmination and counting the number of cells that stained positively for β-galactosidase activity . Both the chtC mutants formed significantly fewer ecmA positive cells than AX4 ( Figure 2C ) . There was no significant change in the proportion of ecmA positive cells when the labeled chtC strains were mixed with either the unlabeled parent or unlabeled AX4 ( data not shown ) . To determine the timing of transdifferentiation , we determined the proportion of ecmA-positive spores formed by the chtC mutants using the [ecmA]:lacZ labeled strains . We found no significant difference compared to AX4 [ecmA]:lacZ cells ( data not shown ) . This finding suggests that in the chtC mutants , a population of prestalk cells that would otherwise have given rise to PST-A cells changes its cell fate and goes on to form spores instead . This process takes place soon after the initial prestalk-cell differentiation - after the induction of tagB expression , but before ecmA induction , a timing that coincides with the peak in chtC mRNA levels ( Figure 1B ) . We further investigated this process by comparing tagB expression levels in 16 h slugs and in fully differentiated spores in both the chtC mutants and in the parental wild type cells ( Figure S3 ) . The level of tagB mRNA was significantly lower in the spores at 24 h as compared to the level in slugs at 16 h , suggesting that the tagB expression observed in the spores of the chtC mutants is not due to a wholesale induction of tagB expression in prespore cells but rather to a transdifferentiation of a small proportion of the prestalk cells . Even though the chtCins mutant had higher levels of tagB expression at 16 h of development ( compared to AX4 ) , the level of tagB mRNA in the spores for both the chtC mutants was not significantly increased compared to a similar AX4 control . These data further support the conclusion that the blue staining observed in spores of the [tagB]:lacZ labeled chtC-mutants reflects transdifferentiation of prestalk cells into prespore cells . Interestingly , even though the PST-A region in the chtC mutant slugs is defective for the expression of two separate markers – tagB and ecmA – the chtC mutants have no overt defects in stalk morphology or function , suggesting that under laboratory conditions , the expression of these markers is not required for proper PST-A cell function . We also tested whether the chtC gene was required to maintain the prespore cell fate , by observing slugs of either AX4 , chtCins or chtCdel expressing the [cotB]:lacZ marker ( cotB is a well-established prespore marker that is expressed exclusively in prespore cells and spores ) [19] . Neither mutant strain expressed the cotB marker in the prestalk region ( Figure S4 ) , suggesting that the chtC mutant cells are not undergoing transdifferentiation from prespore to prestalk cells and that the directional transdifferentiation we observe is not due to a general defect in cell type differentiation . The chtC mutants are defective in the maintenance of the prestalk cell-fate . We hypothesized that this defect in chtC cells would affect prestalk differentiation of AX4 cells in chimera . In order to test this hypothesis , we examined the pattern of AX4 prestalk cells in chimeric populations . We developed mixed populations of 20% AX4/[ecmA]:lacZ cells and 80% unlabeled chtC cells . When mixed with either the chtCins or the chtCdel mutant , the AX4/[ecmA]:lacZ cells were preferentially localized in the PST-A region ( Figure 3A ) . We repeated the experiment using the AX4/[tagB]:lacZ strain [15] , and found similar results ( Figure 3B ) . These experiments were also carried out at a 1∶1 ratio between AX4 cells and the chtC mutants , and qualitatively similar results were observed ( data not shown ) , though the effects were more pronounced at a 1∶4 ratio . To quantify this finding , we mixed AX4/[ecmA]:lacZ cells with each of the chtC mutants , and developed them in chimera . We collected samples at 22 and 24 hours , dissociated the structures and counted the number of cells that stained positive for β-galactosidase activity . The presence of either of the two chtC mutants caused an increase in the number of ecmA positive cells in AX4 ( Figure 3C ) , suggesting that the chtC mutants may cheat by causing an increase in the proportion of AX4 prestalk cells . A simple explanation of these results is that in chimera , a defect in prestalk differentiation in the PST-A region of the chtC mutants is compensated for by AX4 cells , which then occupy the PST-A region to fill the void , and differentiate into more prestalk cells . As such chimeric mixtures complete development , AX4 cells thus form a smaller proportion of spores compared to the chtC mutants , and get cheated upon . In clonal chtC populations , in spite of the defective prestalk marker expression , cells of the chtC mutants take on the PST-A cell-fate and are able to form morphologically normal fruiting bodies , with similar numbers of spores compared to clonal AX4 populations . The model proposed above predicts that the ability of the victim to contribute to the PST-A region is important for the cheating mechanism of the chtC mutants . If the model were correct , the cheating phenotype of the chtC mutants would be correlated with the ability of their chimeric counterparts to contribute to the PST-A region , and consequently differentiate an increased number of prestalk cells . In order to test this prediction , we mixed the chtC mutants with two other mutants that avoid the PST-A region in chimera with AX4 cells , the tagA– and tagB– mutants , and examined prestalk differentiation and spore production . The tagA– and the tagA–/[ecmA]:GFP strains were described previously [20] , [21] . The tagA– mutant has defects in cell-type specification , and does not contribute to the PST-A region and to the terminal stalk structure in chimera with AX4 cells . We examined the patterning of the tagA–/[ecmA]:GFP cells at the slug stage of development . As expected , the tagA–/[ecmA]:GFP cells showed a wild-type like pattern of fluorescence in the anterior part of the slug when developed as a clonal population ( data not shown ) and in 1∶1 mixtures with the unmarked tagA– strain ( Figure 4A a ) . In chimerae with AX4 , the tagA–/[ecmA]:GFP cells showed almost no fluorescence in the prestalk region , consistent with the published observations [20] ( Figure 4A b ) . The results of mixing the tagA–/[ecmA]:GFP cells with either one of the chtC mutants were nearly indistinguishable from that seen when mixing tagA–/[ecmA]:GFP with the wild type ( Figure 4A c , d ) . Since the tagA– prestalk cells do not appear to occupy the PST-A region in chimera with the chtC mutants , we predicted that the proportion of tagA– prestalk cells would also be unaffected in chimerae with chtC . To test this prediction , we mixed tagA–/[ecmA]:GFP cells with either of the chtC mutants in a 1∶1 ratio , developed them and counted the proportion of fluorescently labeled cells after 22 and 24 hours of development . Neither the chtCins nor the chtCdel mutant affected the proportion of [ecmA]:GFP positive cells formed by the tagA– mutant ( Figure 4B ) . Thus the tagA– mutant appears unaffected by the presence of the chtC mutants in chimera , unlike the phenotype seen in the case of wild-type cells ( though it is possible that the lower sensitivity of detection of the GFP reporter as compared to β-galactosidase may be preventing the observation of subtle effects ) . According to our model , these data would suggest that the chtC mutants should not be able to cheat on the tagA– mutant . We performed similar experiments with the tagB– and tagB–/[ecmA]:lacZ strains [14] . The tagB– mutant is unable to proceed beyond the tight aggregate stage of development in a clonal population . However , in chimera with AX4 , tagB– cells can proceed through development , but do not contribute to the PST-A region or to the stalk . We studied the patterning of the tagB–/[ecmA]:lacZ cells at the slug stage of development . As expected , the tagB–/[ecmA]:lacZ cells occupy the PST-O zone when mixed with AX4 cells at a 1∶4 ratio , leaving a substantial portion of the tip ( PST-A region ) unstained ( Figure 5A a ) . However , in 1∶4 chimerae with the chtC mutants , the tagB– prestalk cells were considerably anteriorized , and occupied a larger portion of the PST-A zone ( Figure 5A b , c ) . Similar results were seen at a 1∶1 ratio between the tagB– cells and the chtC-mutants ( data not shown ) , though the phenotype was more pronounced in the 1∶4 chimerae . Based on this observation , our model predicts that the tagB– mutant would differentiate more prestalk cells in chimerae with the chtC mutants . We tested this prediction and observed that in chimerae with the chtC mutants , the tagB–/[ecmA]:lacZ strain produced a higher proportion of [ecmA]:lacZ positive prestalk cells ( Figure 5B ) , similar to the phenotype seen when AX4 is mixed with the chtC mutants . Thus , the tagB– mutant cells behave like the wild type AX4 cells in chimerae with the chtC mutants , suggesting that the chtC mutants would cheat on tagB– cells . We first tested the spore production of the tagA– and tagB– mutants in control chimerae with the wild type AX4 . We grew the strains clonally , mixed each strain at a 1∶1 ratio with AX4 cells and allowed the chimerae to develop . We determined the ratio of spores formed by each strain after development ( Figure 6A ) . In terms of cheating , both the tagA– and the tagB– mutants were neutral when compared to AX4 , each forming approximately 50% of the spores in the 1∶1 mix . We then performed mixing experiments between the chtC mutants and either the tagA– or the tagB– mutants ( Figure 6B ) . We found that neither chtCins nor chtCdel cheated on the tagA– mutant , but both cheated on the tagB– mutant . These results correlate well with the effects of the chtC mutants on the prestalk differentiation of the tagA–and the tagB– mutants in chimerae with chtC , thus supporting our hypothesis .
The chtC mutants undergo a transformation of cell-fate , since cells with a prestalk history form spores . This is coincident with a PST-A specific defect in the expression of prestalk markers such as tagB and ecmA , suggesting that cells fated to occupy the PST-A region transdifferentiate and form spores instead . Thus the chtC gene appears to be involved in the maintenance of the PST-A cell-fate . This idea is also supported by the cell-type specificity of chtC gene expression , since during late development , chtC is the most stalk-enriched gene described to date , being expressed predominantly in the stalk , and not in other prestalk-derived tissues like the upper and lower cups . Thus , chtC is one of the few genes identified to be involved in maintaining cell-fate [21]–[23] . It is interesting to note that despite the defects in maintenance of the prestalk cell-fate and expression of prestalk markers , stalk morphology and function in the chtC mutants appears indistinguishable from that of the wild type . This finding raises the question of why the chtC mutants have not spread within the population , and why the chtC gene still exists in the genome in Dictyostelium . It is possible that the chtC mutants have fitness defects in growth or development in the wild , or under specific environmental conditions that we have not explored in the laboratory . Additionally , it has been shown that mutants that can resist cheating by the chtCins mutant can be selected for in a population containing the chtCins mutant [10] . Such cheater-resistors can even inhibit the cheating by the chtC mutants , and may thus help to maintain the chtC gene in the population [10] . The chtC mutants differentiate a population of cells that express prestalk markers , but adopt the prespore cell-fate . This transdifferentiation is associated with a decrease in the number of cells that express the late prestalk marker ecmA . In chimerae between AX4 and chtC cells , the AX4 cells differentiate a higher number of ecmA-positive cells . The simplest explanation for these observations is that the void in prestalk cells in chtC is detected by the AX4 cells , which then compensate by differentiating more prestalk cells . The proportions between prestalk and prespore cells are almost constant in Dictyostelium slugs over a wide range of total cell numbers , indicating that well-regulated proportioning mechanisms control the initial differentiation of prestalk and prespore cells [24] . Our data support the hypothesis that there is a feedback mechanism that helps to sense the proportions of properly differentiated prestalk cells , and regulates the differentiation of as yet undifferentiated cells into the required cell-types as development proceeds . The presence of the chtC mutants in chimerae affects the prestalk differentiation and patterning of the wild-type cells , which is likely to be the direct mechanism of cheating . In order to test whether prestalk patterning was important for cheating , we utilized two other prestalk differentiation mutants - tagA– and tagB– . In both cases , the ability of the chtC cells to affect patterning was directly correlated to the cheating behavior , suggesting that the patterning was indeed important for cheating . Nevertheless , the ability to cause wild-type cells to occupy the PST-A zone in chimera does not necessarily equate to cheating , since neither tagA– nor tagB– are cheaters . In chimerae , tagA– mutants also cause wild-type cells to occupy the PST-A region and to be the sole contributor of stalk cells [20] , but the tagA– mutants are not cheaters . This finding suggests that the mechanism of cheating by the chtC mutants is more than a passive recognition of a PST-A cell deficiency by the wild-type members of the chimerae . The mechanism of cheating seen in chtC is significantly different from that of chtA ( fbxA ) [8] . Though chtA is an obligatory cheater that is unable to form spores in clonal populations , it differentiates a higher proportion of prespore cells in slugs [8] . The presence of wild-type cells rescues its development , allowing it to differentiate a higher number of spores in chimeric fruiting bodies . On the other hand , even though the chtC mutant has defects in cell-fate maintenance , it is morphologically normal and does not require the presence of wild-type cells to complete development , yet it ends up forming more than its fair share of spores in chimerae with wild-type cells . Furthermore , while both chtA and chtC increase the prestalk differentiation of their victims , chtA causes the victim's prespore cells to transdifferentiate into stalk cells [8] , whereas chtC causes a higher number of the victim's cells to initially differentiate as prestalk cells . These observations suggest that chtC might be affecting wild-type differentiation earlier than the chtA mutant . The tagB– mutant is morphologically rescued when mixed with AX4 cells , and goes on to complete development , although tagB– cells do not contribute to the PST-A region in the chimerae [14] . However , when mixed with the chtC mutants , tagB– cells become anteriorized and occupy most of the PST-A region , except for the very tip . This finding suggests that the presence of the chtC mutants partially overcomes the tagB– defect . It is therefore likely that the chtC mutants affect their chimeric partners before the tagB gene acts , in the sequence of developmental events . Since the very tip of the slug does not contain tagB– prestalk cells ( unlike AX4 ) , it is also likely that tagB– cells are defective in forming several prestalk cell types , and the defect in contributing cells to the very tip of the slug is separate from the PST-A cell defect . The tagA– mutant , on the other hand , is unaffected by the chtC mutants in chimerae , suggesting that the chtC gene functions later than tagA , and consequently the chtC mutants are unable to affect the tagA– cells . These suggestions are consistent with the timing of expression of the three genes - both tagA and chtC are expressed throughout development , but their expression peaks at 2 and 12 hours respectively [21] . The tagB gene is first induced much later , at about 8 hours , and peaks at 20 hours of development [14] . Both the tagA– and tagB– mutants have defects in prestalk differentiation , similar to the chtC mutants , and both have morphological defects in stalk formation . It has been suggested that the wild type preferentially forms PST-A cells in chimera with these mutants since the mutants are defective in forming those cells [20] . A similar explanation can account for the finding that wild-type cells preferentially contribute to the PST-A region in chimerae with the chtC mutants . Though the chtC mutants do not appear to be functionally defective in stalk formation , they are defective in the expression of prestalk markers . This observation supports the hypothesis that cells with appropriate expression of prestalk genes contribute preferentially to the stalk ( especially the PST-A region ) , possibly as a form of stalk “quality-control” . The chtC mutants appear to be taking advantage of this PST-A “check-point” . Their presence in chimeric mixtures induces the wild-type cells to form stalk cells even though the chtC mutants have the ability to do so themselves , and this leads to an increase in their own spore production at the expense of their victim . This is thus an example of developmental cheating where in the presence of a genetically distinct strain , a cheater mutant is subverting a developmental pathway to increase its own fitness . Microbial social behaviors are broadly divided into two categories [25] – the production of public goods , and the formation of fruiting bodies as seen in Dictyostelium and Myxococcus xanthus . While the former is normally concerned with a single ( biosynthetic ) pathway , the latter may involve various signaling pathways that normally lead to complex developmental processes . Consequently , developmental processes are likely to be manipulated for cheating in these social systems , similar to that seen in super-organisms like social insect colonies . We see an example in this study , where a cheater mutant is manipulating an existing developmental pathway of cell-fate determination and proportioning to exploit other clones . The cooperative system in Dictyostelium thus offers a good opportunity to study developmental cheating mechanisms at the genetic and cellular level .
The D . discoideum strains used in this study are described in Table 2 . The chtCins strain was described before as the chtC mutant [10] . To generate the chtCdel strain , we amplified two fragments from the knockout vector by PCR ( Upstream arm primers: 5′-CTTGACATGCGAAATGGC-3′ , 5′-GAAGGGACTCCATAAGTATGAG-3′; downstream arm primers: 5′-GTCTTCCAGATGAAAGTTGC-3′ , 5′-CCTAATGCAGCACATACTGC-3′ ) . The PCR fragments were cloned between the KpnI and ClaI sites of the pLPBLP plasmid , and the entire plasmid was used as a knockout construct to delete most of the endogenous chtC gene . For both the chtC mutants , the BSR cassette was subsequently removed by transforming the cells with the pDEX-NLS-Cre plasmid [26] . We also created a Cre-expressing plasmid carrying the hygromycin-resistance cassette to use in strains that are already G418-resistant . We transposed the tetr-hygr cassette from the EZTN::tetr-A15hygr plasmid ( a kind gift from J . Williams ) into the pDEX-NLS-Cre plasmid ( just downstream of the act8 terminator ) to generate the pDEX-Cre-hygr plasmid . The tagB– mutant was generated by transforming the ptgB-BSR plasmid ( a kind gift from W . F . Loomis ) into AX4 . ptgB-BSR is a ClaI-rescued plasmid from a REMI insertion of the pBSRdelBglII plasmid into position 2672 of the tagB coding region . The chtCins mutation was generated in the AX4/[cotB]:lacZ ( TL1 ) and AX4/[ecmA]:lacZ ( TL6 ) strains , and the BSR cassette was subsequently removed by transforming cells with the pDEX-Cre-hygr plasmid , to give the chtCins/[cotB]:lacZ and chtCins/[ecmA]:lacZ strains respectively . To create the chtCdel/[cotB]:lacZ , chtCdel/[ecmA]:lacZ , chtCins/[tagB]:lacZ and chtCdel/[tagB]:lacZ strains , we transformed the pSP70-LacZ [19] , p63NeoGal [27] or the ptagB/lacZ [15] plasmids into the respective chtC mutants . D . discoideum cells were grown in suspension cultures in HL5 [28] with the necessary supplements . All strains were grown in HL5 medium without antibiotics for 24–48 hours prior to setting up any experiments , to avoid the potential effects of antibiotics on cell behavior . One labeled strain from each background was mixed with AX4-GFP cells to test the effect of the antibiotic on mixing experiments ( Figure S5 ) . Plasmid transformation was carried out essentially as described earlier [29] , with the following modifications: cells were resuspended at a final density of 3×107 cells/ml before transformation , electroporated twice , and the transformants were recovered in HL5 with 10% fetal bovine serum for 24 hours prior to the addition of drugs . Depending on the plasmids , transformants were selected with either Blasticidin S ( 10 µg/ml ) or G418 ( 5 µg/ml ) . Transformants were grown clonally on SM-agar plates in association with K . aerogenes [28] , and then re-tested for drug resistance in 24 well-plates containing HL5 with the drug . When appropriate , drug-resistant clones were tested for the correct recombination event by PCR and by Southern blot analysis . We developed cells as described earlier [29] with the following modifications: cells were washed with KK2 buffer ( 16 . 3 mM KH2PO4 , 3 . 7 mM K2HPO4 , pH 6 . 2 ) , resuspended at a density of 1×108 cells/ml , and 5×107 cells were deposited on each nitrocellulose filter . For the mixes , the cells were grown separately , and mixed before development . We collected all the cells ( after 40–48 hours ) , treated them with 0 . 1% NP40 to select for spores , and , in the case of GFP-labeled strains , we counted them as described [29] . For mixes with tagB– , the spores were plated out clonally on SM-agar plates in association with K . aerogenes [28] , and the plaques were scored by their developmental morphology . For the mixes between the rest of the mutants , spores were plated out similarly , and cells from individual plaques were transferred to two 96-well plates in HL5 containing 10 µg/ml Blasticidin S , and scored for drug-resistance . For the sporulation efficiency experiments , cells were developed as above , and all cells were collected after 40–48 hours of development . NP40-resistance was calculated as the ratio of the number of visible spores after NP40-treatment to the same number prior to NP40-treatment . Sporulation efficiency was calculated as the ratio of the NP40-resistant spores obtained to the number of cells originally plated . These spores were then plated out clonally on SM-agar plates in association with K . aerogenes [28] and germination efficiency was calculated as the ratio of the number of plaques obtained to the number of spores plated . For fluorescence microscopy of developing structures with tagA–/[ecmA]:GFP cells , the cells were developed on KK2 plates as described [9] . For the segregation assay , we labeled cells with either CellTracker CMFDA or CellTracker Orange CMRA ( Molecular Probes ) as described [29] . After labeling , we mixed cells from the appropriate strains at a 1∶1 ratio and a final density of 5×106 cells/ml . We then spotted 40 µl of this cell suspension on KK2 ( non-nutrient ) agar plates , allowed the cells to develop for 8 hours , and then photographed with both transmitted light and fluorescence microscopy . The fluorescence images were overlaid and are shown as color photographs . Developing structures were fixed and stained in situ with X-gal ( for β-galactosidase activity ) as described previously [18] , and were counterstained with 0 . 02% eosin Y [30] . For each experiment , tens of structures were observed in at least 2 independent biological replications , and representative structures are shown in the figures . Staining of dissociated cells was done essentially as described earlier [18] , except that the developing structures were passed through an 18G1½ needle , and treated with pronase ( 0 . 1% pronase , 0 . 1% β-mercaptoethanol , 150 mM NaCl , 50 mM Tris pH 7 . 0 ) for 10 minutes at room temperature for efficient dissociation . GFP-labeled cells were counted directly after dissociation using phase-contrast and fluorescence microscopy . For slug migration , cells were washed twice with double-distilled water , and 108 cells were streaked on 2% Agar-Noble plates made with double-distilled water . The plates were incubated in a dark chamber with a unidirectional source of light for 48 hours , and exposed to overhead light to induce culmination . Developing structures were collected after migration and stained as above . Spore staining was carried out as described previously [18] . Genomic DNA was prepared as described earlier by the CTAB method [31] . Southern blot analysis was performed by standard methods [32] . RNA extraction and Northern blot analysis were performed as described previously [33] . The blots were hybridized with DNA probes made by random-primer labeling [34] . We used a PCR fragment from pLAS5 [9] to probe for the chtC gene . The abundance of the chtC mRNA was determined by Q-RT-PCR as described , using rnlA ( Ig7 ) to normalize for cDNA levels [35] . The primers used were: chtC 5′-TTCACCAAATCCACTAGACTGTC-3′ and 5′-CAGTTGCTTTCTTACGTGCAAG-3′and Ig7 5′-TTACATTTATTAGACCCG AAACCAAGC-3′ and 5′-TTCCCTTTAGACCTATGGACCTTAGCG-3′ . The abundance of the tagB mRNA was also similarly determined by Q-RT-PCR ( primers: 5′-TTTCCCAACTGGCGAATC-3′ and 5′-CCTAAACCACCGATACCAATC-3′ ) . In situ RNA hybridization was done as described [36] with the following modifications: hybridization was done in the same solution as the pre-hybridization; both steps , as well as washing were done at 50°C , and the final wash was done in 0 . 1X SSC . A digoxigenin-labeled RNA probe was made by in vitro transcription from the plasmid pLAS5 using the T7 promoter with the DIG RNA labeling kit from Roche . | Cooperative systems are susceptible to exploitation by cheaters who enjoy the benefits of cooperation without paying the costs . Such conflict is seen in biological systems at every level from individual genes within a cell to individuals within societies . The social amoebae Dictyostelium discoideum have a unique cooperative system in which large numbers of individual cells aggregate to form fruiting bodies with reproductive spores , and dead stalk cells that may help the survival and dispersal of the spores . Fruiting bodies can contain several genotypes , and hence can be exploited by cheater cells that preferentially form spores without contributing fairly to the stalk . We have studied a mutant , cheater C ( chtC ) , which is defective in forming certain stalk cells , but is still able to form fruiting bodies on its own . However , when wild-type cells are mixed with chtC cells , the wild-type cells compensate for the stalk-forming defect of chtC and form more of the stalk cells . In that way , chtC cells cheat by taking advantage of developmental processes that normally regulate cell-type proportions . This study shows that existing mechanisms of developmental regulation can be exploited by cheater mutants , and the social amoebae offer a good system to study such mechanisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/cell",
"differentiation",
"developmental",
"biology/developmental",
"evolution",
"evolutionary",
"biology/developmental",
"evolution"
] | 2010 | Cheating by Exploitation of Developmental Prestalk Patterning in Dictyostelium discoideum |
Evaluating the effectiveness of malaria control interventions on the basis of their impact on transmission as well as impact on morbidity and mortality is becoming increasingly important as countries consider pre-elimination and elimination as well as disease control . Data on prevalence and transmission are traditionally obtained through resource-intensive epidemiological and entomological surveys that become difficult as transmission decreases . This work employs mathematical modeling to examine the relationships between malaria indicators allowing more easily measured data , such as routine health systems data on case incidence , to be translated into measures of transmission and other malaria indicators . Simulations of scenarios with different levels of malaria transmission , patterns of seasonality and access to treatment were run with an ensemble of models of malaria epidemiology and within-host dynamics , as part of the OpenMalaria modeling platform . For a given seasonality profile , regression analysis mapped simulation results of malaria indicators , such as annual average entomological inoculation rate , prevalence , incidence of uncomplicated and severe episodes , and mortality , to an expected range of values of any of the other indicators . Results were validated by comparing simulated relationships between indicators with previously published data on these same indicators as observed in malaria endemic areas . These results allow for direct comparisons of malaria transmission intensity estimates made using data collected with different methods on different indicators . They also address key concerns with traditional methods of quantifying transmission in areas of differing transmission intensity and sparse data . Although seasonality of transmission is often ignored in data compilations , the models suggest it can be critically important in determining the relationship between transmission and disease . Application of these models could help public health officials detect changes of disease dynamics in a population and plan and assess the impact of malaria control interventions .
Evaluating the effectiveness of malaria control interventions on the basis of their impact on transmission is increasingly important as countries consider elimination as well as malaria control . However , direct measurement of transmission , such as by the entomological inoculation rate ( EIR ) ( a measure of human exposure defined by the number of infective mosquito bites per human in a given time period ) , involves mosquito capture . This is extremely labor-intensive , and is only reliable in high transmission areas and seasons [1] . In areas of low transmission , or during dry seasons , identifying a sufficient number of sporozoite-positive mosquitoes makes this exercise excessively time- and resource-intensive , often precluding collection of a full year's worth of data and making estimates of seasonality challenging . Alternatives are to estimate transmission rates from sero-conversion rates [2] , [3] or by calculating force of infection ( FOI ) from combining information on prevalence and treatment [4] . Estimating both the exposure to infectious mosquitoes and subsequent FOI from parasite prevalence in areas of high transmission is difficult due to superinfection and immunity . Mathematical models are useful in examining relationships between malaria indicators , allowing translation of routine health center data into measures of transmission and addressing concerns with previously implemented methods of measuring transmission [5] . Understanding the seasonal pattern of malaria transmission is important for planning control interventions , for example the timing of deploying indoor residual spraying ( IRS ) and seasonal malaria chemoprophylaxis ( SMC ) which are implemented ahead of the peak transmission months . Given the wide range of seasonal patterns combined with transmission intensities that exist in areas of the world with malaria transmission , and due in large part to the absence of robust field data , the effect of seasonality on the relationship between malaria indicators has not been studied in great detail . Attempts have been made to define [6] , [7] and quantify [8] the relationship between seasonally varying covariates and transmission based on available studies on malaria transmission and disease burden , but results for the latter were only found to be reliable in areas of very high transmission ( EIR>100 infectious bites per person per year ) [6] . One approach for quantifying transmission in areas without EIR data is to use simulation models to analyze how different malaria indicators ( parasite prevalence , prevalence of uncomplicated and severe episodes , mortality ) relate to each other , and how they relate to transmission as measured by EIR [5] . To validate such models , a straightforward approach would be to compare the simulated relationships between indicators to those observed in the field . However , when relationships between indicators differ in places with disparate patterns of seasonality , such an approach becomes challenging . This study uses simulation models to analyze whether relationships between malaria indicators are likely to vary by intensity and pattern of seasonality . Analysis of these simulation results can help identify the best way of quantifying transmission for the purposes of specifying the seasonal patterns to drive existing models of Plasmodium falciparum dynamics . This in turn will assist in planning for malaria control by allowing for the selection of interventions tailored to the level of transmission in a given location , and monitoring the effectiveness of those interventions by their impact on transmission .
This experiment utilizes an ensemble of simulation models of transmission of malaria developed by a team at the Swiss Tropical and Public Health Institute ( Swiss TPH ) and Liverpool School of Tropical Medicine . These models form part of the OpenMalaria platform that makes the considerable code base written in C++ accessible to the public through an online wiki [9] . Based on a stochastic series of parasite densities for individual infections , stochastic individual-based models of malaria in humans [10]–[12] are linked to a periodically-forced model of malaria in mosquitoes [13] in order to simulate the dynamics of malaria transmission and the impact of intervention strategies for malaria control . Details of the methods to create and parameterize the transmission model used in this project have been previously published [10]–[13] and therefore are not covered in this paper . Models are fitted to 10 objectives using 61 standard scenarios as described in Smith et al . 2008 [11] . The transmission model is calibrated by the seasonal pattern of the EIR with units of infectious bites per person per year . Simulations were run for one human life span to induce a stable level of immunity in the population . Each simulation was repeated on an ensemble of 14 model variants with varying assumptions on mass action , heterogeneity of exposure , decay of acquired immunity , co-morbidities , and access to treatment as described in Smith et al [12] to address model uncertainty , with five random seeds to address stochasticity . The overall objective of estimating transmission in areas without EIR data was addressed by applying the OpenMalaria modeling platform to simulate malaria with different levels of transmission and patterns of seasonality observed in malaria-affected locations , and deriving outputs for all other malaria indicators . Table 1 describes the indicators chosen as simulation outputs that were evaluated in this study . Relationships between all indicators for the different values of EIR and different seasonality profiles were estimated from simulation results ( described below ) using Stata v12 ( College Station , TX ) . For this process the indicators were calculated for the whole population , with the exception of the relationships involving mortality which were limited to children under five due to a lack of data in older age groups for validation purposes . The baseline scenario used in these experiments was based on a scenario previously parameterized for the Rachuonyo South district in the highlands of western Kenya [14] . The model assumes no interventions beyond case management through the health system as described in Tediosi et al . [15] , a main vector of A . gambiae s . s . , and artemisinin combination therapy ( ACTs ) as the first line antimalarial . Simulations were run on a population of 100 , 000 individuals over three years with monthly surveys of malaria outcomes . To quantify the “amount” of seasonality in a location a seasonality index ( φ ) was defined in order to describe the variations in transmission within one year in a given location . The methodology presented here is general and can be used for any measure of transmission , but the example below is used with EIR . We let T denote the period ( 1 year ) and let f ( t ) be a positive continuous periodic function that denotes transmission at time t , with f ( t+T ) = f ( t ) >0 for all t≥0 . The mean level of transmission ( over 1 year ) is , In a similar manner to the coefficient of variation in statistics , we define φ as the normalized square root of the integral of the squared difference between f ( t ) and its mean , This seasonality index , φ , allows us to quantify the level of seasonality of transmission in a given location with one positive real number , differentiating between “amounts” of seasonality for transmission patterns with the same number of peaks . Because malarious areas in general have either one or two peak transmission seasons , there could be seasonality patterns in different locations that lead to the same seasonality index , . We therefore label the seasonality profile with both the seasonality index and the number of peaks . The simulations described here treat transmission in the absence of interventions as periodic with a one year period [13] . One scenario with a seasonality pattern of constant annual transmission ( = 0 ) and five scenarios with varying seasonal transmission patterns ( = 1 , one peak; = 1 , two peaks; = 0 . 5 , two peaks; = 2 , one peak; = 2 , two peaks ) were created , described in Table 2 and Figure 1 . These six patterns were chosen to represent the range of seasonal patterns of malaria transmission existing in the malaria endemic world , namely because there are usually not more than two peak transmission seasons . The seasonality profiles with = 2 exhibit large variations in seasonality . For = 2 with one peak , 86% of annual transmission is focused in the three peak transmission months , while for = 2 with two peaks , the peak is narrower with 95% of annual transmission occurring in the three months of the higher peak . The results of what this means for prevalence and morbidity over one year can be found in Figure S1 in Text S1 . Seasonality patterns were repeated for eleven values of annual average EIR from 0 . 5 to 365 . Complete details of the methods behind the experiment creation can be found in Text S1 . The relationships between malaria indicators were estimated using fractional polynomial regression as described in more detail in Text S2 . In order to gauge the model's ability to reproduce field data , a validation exercise was completed by comparing simulation results to data not used in the original process of model fitting from previously published studies . The relationships for validation , the datasets used and how they relate to model fitting are described in Table S1 . While the annual average EIR in the scenarios used for estimating the relationships between malaria indicators were capped at a value of 81 . 4 , scenarios for validation were simulated up to an average of 365 infectious bites per person per year . This tailors the analysis to low- to mid-range values of annual average EIR where this tool will be the most applicable , while still allowing for a more comprehensive range of annual average EIRs that appear in the validation datasets .
When analyzing the relationship between EIR and other malaria indicators , the differences between seasonality profiles are greatest at moderate levels of EIR ( Figure 2a–d ) . Results are similar between seasonality profiles at both ends of the EIR spectrum for uncomplicated and severe disease , but seasonality impacts the relationship with prevalence and mortality more at higher values of EIR ( Figure 2a–d ) . The Beier et al . dataset , describing the relationship between EIR and parasite prevalence in children under five in sites across Africa , has been applied for a previous validation of the OpenMalaria model [16] . One site out of 31 as published separately was used to fit the model for incidence of asexual blood stage infection , as indicated in Table S1 . Compared to the results presented in Beier et al . [17] , simulation results are within the range of observed values for low and medium values of EIR , but predict a slightly lower prevalence in extremely high EIR settings , especially in a setting with no seasonality ( Figure 3 ) . Perhaps this is because observed results reach up to 1 , 000 infectious bites per person per year while the simulated scenarios were capped at 365 . While the observed relationship is fitted as log-linear , the simulated relationship starts levelling off at an EIR of 100 . The relationship between parasite prevalence and uncomplicated episodes is non-monotonic ( Figure 4a ) for all values of . It can be noted that the simulated relationship between parasite prevalence and severe disease shows more stochasticity than the other relationships with parasite prevalence in areas of lower prevalence ( Figure 4b ) . This variation can be attributed to model uncertainty , in particular differing assumptions about access to treatment , rather than to the effect of seasonality . For uncomplicated disease , severe disease and mortality , the effect of seasonality is greater in areas of higher parasite prevalence; the variation increases once prevalence reaches 40% ( Figure 4a–c ) . Compared to the results presented in Okiro et al . [18] the model is able to reproduce the general pattern of the relationship between severe pediatric malaria and prevalence in children aged 2–10 in children under 1 year as well as in children aged 5–9 , with the burden of malaria moving to older age groups as prevalence is reduced ( Figure 5 ) . Compared to the results presented in Korenromp et al . [19] , which describes the relationship between parasite prevalence and both malaria-specific and all-cause mortality in children under 5 , the model is able to capture the general pattern for the relationship between malaria-specific mortality in children under five for low and moderate prevalence settings ( Figure 6 ) . There appears to be variation across sites in the observed data that may be explained by the ability of verbal autopsy to capture indirect deaths due to malaria in different settings [20] . Nine sites ( for which EIR estimates were available ) out of the 28 sites included in the study were used to fit the model of direct malaria mortality in relation to EIR , as indicated in Table S1 . At lower numbers of uncomplicated episodes per person per year , seasonality does not play a role in the relationship with severe episodes ( Figure 7 ) . The curves separate at levels above 1 . 25 uncomplicated episodes per person per year with two-peak scenarios = 1 and = 2 diverging from the other values of ( Figure 7 ) . The scatter plot of simulation results showed no discernible relationship between mortality and either uncomplicated or severe episodes , and are therefore not shown here . Age prevalence curves are validated by comparing simulation results to those presented in Carneiro et al , which report on the age distribution of children with clinical malaria , hospital admissions with malaria and malaria-diagnosed mortality for different categories of intensity and seasonality of malaria transmission identified from a systematic review epidemiological studies [6] . It should be noted that there are differences in the classification of degree of seasonality between the observed and simulated data . Carneiro and colleagues describe settings with marked seasonality as those with greater than or equal to 75% of episodes concentrated less than or equal to 6 months of the year . In the OpenMalaria simulations , marked seasonality is defined as the setting with = 2 . The reported estimated median ages and inter-quartile ranges ( defined as the 50th percentile of the best-fitting distribution for each outcome and transmission scenario ) from these fitted models for each level of transmission and level of seasonality are compared to estimates from fitted OpenMalaria simulation results to validate age prevalence curves of the malaria indicators mentioned above . In all cases , the results of the OpenMalaria simulations are comparable to the previously published results ( Figure 8 ) .
Due to the lack of understanding of the relationship between EIR and other malaria indicators based on challenges in measuring EIR from entomological studies , modeling is able to further define the relationships between indicators and help clarify details of what cannot measured from field studies but is nonetheless necessary knowledge about malaria indicators . This is of value for malaria control program managers because it provides insight on transmission without substantial field studies . These models can be used to simulate the likely range of values in areas without access to adequate field data . Empirical studies of the relationships between different malaria indicators are challenging because these relationships may in principle be affected by many , often poorly characterized , contextual factors , with the degree of seasonality being possibly one of the most important . The original fitting of the OpenMalaria model parameters to multiple field datasets used a standard pattern of seasonality of transmission from Namawala , Tanzania; effects of seasonality observed in these results are thus not an artifact of the fitting process . Simulations suggest that with equal levels of average annual transmission , the level of seasonality , i . e . whether malaria transmission is fairly constant over the course of a year versus peaks in certain months , affects the relationship between malaria indicators . An increase in the degree of seasonality has a greater impact on outcomes with moderate levels of EIR and prevalence . There is greater stochasticity in simulation results for scenarios with higher amplitude of the annual cycle compared to scenarios with a constant level of transmission . There have been previous attempts to create a measure for the seasonality of malaria transmission [21]–[23] , mainly relying only on rainfall and/or vector abundance to describe the proportion of transmission occurring within a certain number of months . The approach to developing the seasonality index presented here is in response to the need to provide a quantitative metric for differences between seasonal patterns . Results indicate that this index does not distinguish well between patterns that have a different number of peaks ( Figure 2 ) ; therefore the number of peaks should also be noted in any analysis of studies that employ this index . Areas with seasonal malaria transmission typically have substantial variation in rainfall and transmission with numerous small peaks , but normally only have one or two main seasons . The total number of peaks can thus be assumed to be limited to a maximum of two . The difference in results for different patterns within the same seasonality index calls into question the assumptions behind the drivers of the relationships between malaria indicators . Scenarios with a higher degree of seasonality , regardless of number of peaks , return lower levels of prevalence , disease and mortality for a given level of transmission . An important driver is multiple concomitant events; when two illness episodes occur at the same time they are only considered as one , which may occur more frequently in high seasonality scenarios . At more mild patterns of seasonality , this phenomenon is only seen at higher levels of transmission . These results also potentially indicate an effect on acquisition of immunity in these settings , a consideration when modeling the relationship between transmission and the acquisition of immunity in a population . Several model variants differ in their assumptions about immunity [12] , and while outside the scope of this paper , an important question for future investigation would be the impact of this aspect of the models variants and the effect , if any , that occurs for different seasonal patterns of transmission . Results indicating the impact of seasonality on the relationship between malaria indicators is relevant to malaria epidemiology and control because , as has been described in Carneiro et al [6] , areas with similarly high average annual prevalence result in less frequent cases of malaria in highly seasonal settings . A focused empirical analysis of this effect would be another welcome addition to the understanding of the subject . Access to treatment has the potential to impact the relationships between transmission and other malariological indicators such as severe disease and mortality . The higher the proportion of malaria cases that are treated with effective antimalarials the more the parasite reservoir in the human host population is suppressed , the fewer gametocytes are available , and the less likely it is that mosquitoes are infected . The authors are not aware of any empirical studies of the relationship between access to treatment and population-level health outcomes . However , recent work by Briët and Penny investigates the impact of access to treatment on the OpenMalaria model [24] . The relationships between severe episodes and other indicators ( Figures 2c , 4b , 7 ) may depend more on access to effective case management , indicated by the variance in simulation results which is due to model uncertainty rather than the effect of seasonality . There are direct implications on control programs for the relationship between seasonality and the expected number of uncomplicated cases for a given level of parasite prevalence . Locations with poor monitoring and surveillance systems resulting from complex emergencies or insufficient reach of the public sector may have readily-available parasite prevalence data as a result of research activities . These results may impact how routine data from the case management system in these locations are able to be used to inform study design for the implementation of seasonality-dependent interventions such as IRS and SMC . Two sources mentioned in this model validation were also used in the original model fitting [12] . However , as indicated in the Results section and in Table S1 , the relationships used here for validation were not the same relationships ( Korenromp et al . ) or subsets of data ( Beier et al . ) used for fitting . Although both help parameterise the model , because this process was independent to the relationships being validated , they can therefore be treated as available for validation . Each simulation result is a point in multidimensional space with each dimension corresponding to one malaria indicator . However , to determine the relationship between any two indicators , all simulation points are projected onto a two-dimensional space where the relationship is estimated through fractional polynomial regression . Due to this projection , when two indicators have a monotonically-increasing relationship with a third indicator , they may not necessarily have a monotonically-increasing relationship with each other . For example , while simulated parasite prevalence and mortality both increase with increasing annual average EIR , the same effect will not necessarily be seen on mortality in conditions of increasing prevalence . Similarly , the effects of seasonality appear to decrease as EIR increases , but increase as prevalence increases . While the range of transmission levels and patterns represented in this study are designed to cover a large proportion of malaria endemic areas , there are areas with contexts that will fall outside the scope of this work . There remain areas with extremely high transmission beyond an annual average EIR of 81 . 4 at which this analysis is capped , but these programs are unlikely to be at a stage of malaria control to benefit from applying the methods described in this paper for fine-tuning malaria control interventions as vector control interventions can be effectively utilized to substantially reduce malaria transmission to moderate levels and transmission can be adequately measured with entomological methods . Simulated results were limited to annual average EIR values greater than 0 . 5 . In very low transmission settings infections are sporadic and could be better captured with epidemic models . At very low annual average transmission rates malaria can be sustained by regular importation or the presence of hotspots . The relationships between malaria indicators then depend critically on the degree of transmission heterogeneity and interactions between sub-populations . In these settings , estimating transmission through using serology to estimate EIR or force of infection may be more suitable . Although not currently available in the OpenMalaria transmission model , force of infection and serology will be important components to add to future versions to better simulate the current practice of measuring transmission at low values of EIR . With the inclusion of these indicators , the new model can be calibrated with data on incidence but validated with other indicators ( i . e . prevalence or serology ) . Because of the strong effect of seasonality on the relationships between malaria indicators , it follows that obtaining accurate estimates of transmission across a range of seasonal patterns , not just transmission intensities , is critical for tailoring malaria control and elimination programs to specific country contexts . An accurate map describing seasonal patterns of transmission to attach to maps of transmission intensity and other indicators would be a useful tool . While obtaining this information may not be straightforward , there is a need for research studies designed with measuring not only transmission but also other malaria indicators to ensure the annual pattern of transmission is accounted for . Therefore , goals for reduction in transmission and burden of disease can be further tailored to specific sites . The methods described here will be able to be compiled into a lookup tool that will allow malaria control professionals to enter the data they have on one index and see the range of likely results for other measures of malaria . In addition to estimates , an essential requirement would be providing a means to display the uncertainty of simulation results . Examples of how this might be achieved are discussed in Text S3 and shown in Figures S2–S5 in Text S3 . Such a tool could aid in the planning process of tailoring malaria control interventions to the appropriate level of transmission . | While malaria is still a major public health problem in many parts of the world , control programs have greatly reduced the burden of disease in recent years and many countries are now considering the goal of elimination . Unfortunately , malaria transmission becomes more difficult to measure when it is low because traditional methods involve capturing mosquitoes; an expensive and time-consuming technique . To measure transmission in areas without adequate field data , we run simulations of a mathematical model of malaria over a range of transmission intensities and seasonal patterns to examine how different measurements of malaria ( prevalence , clinical disease , and death ) relate to each other , how they relate to transmission , and if the relationships are likely to vary by seasonal pattern of transmission . These simulated relationships allow us to translate easily measured data , such as clinical case incidence seen at health facilities , into estimates of transmission . This technique can help public health officials plan and assess the impact of malaria control interventions , even in areas without intensive research activities . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"computer",
"and",
"information",
"sciences",
"medicine",
"and",
"health",
"sciences",
"population",
"modeling",
"biology",
"and",
"life",
"sciences",
"infectious",
"disease",
"modeling",
"computational",
"biology",
"computerized",
"simulations",
"malaria",
"parasitic",
"diseases"
] | 2014 | Seasonally Dependent Relationships between Indicators of Malaria Transmission and Disease Provided by Mathematical Model Simulations |
Tetherin ( Bst2/CD317/HM1 . 24 ) is an interferon-induced antiviral host protein that inhibits the release of many enveloped viruses by tethering virions to the cell surface . The HIV-1 accessory protein , Vpu , antagonizes Tetherin through a variety of proposed mechanisms , including surface downregulation and degradation . Previous studies have demonstrated that mutation of the transmembrane domains ( TMD ) of both Vpu and Tetherin affect antagonism , but it is not known whether Vpu and Tetherin bind directly to each other . Here , we use cysteine-scanning mutagenesis coupled with oxidation-induced cross-linking to demonstrate that Vpu and Tetherin TMDs bind directly to each other in the membranes of living cells and to map TMD residues that contact each other . We also reveal a property of Vpu , namely the ability to displace Tetherin from sites of viral assembly , which enables Vpu to exhibit residual Tetherin antagonist activity in the absence of surface downregulation or degradation . Elements in the cytoplasmic tail domain ( CTD ) of Vpu mediate this displacement activity , as shown by experiments in which Vpu CTD fragments were directly attached to Tetherin in the absence of the TMD . In particular , the C-terminal α-helix ( H2 ) of Vpu CTD is sufficient to remove Tetherin from sites of viral assembly and is necessary for full Tetherin antagonist activity . Overall , these data demonstrate that Vpu and Tetherin interact directly via their transmembrane domains enabling activities present in the CTD of Vpu to remove Tetherin from sites of viral assembly .
Tetherin is an antiviral protein that can inhibit the release of a broad-spectrum of enveloped viruses , including retroviruses [1]–[7] , filoviruses [4] , [8]–[10] , arenaviruses [9] , [10] , rhabdoviruses [11] and herpesviruses [6] , [12] . The Tetherin protein is a type-II single-pass transmembrane domain ( TMD ) protein that is also appended with a C-terminal glycophosphatidylinositol ( GPI ) moiety as a second membrane anchor [13] . Between the membrane anchors is a coiled-coil ( CC ) domain that is covalently linked to a second Tetherin molecule by three intermolecular disulfide bonds [14]–[18] . Both membrane anchors and the CC domain are necessary for activity , and the membrane anchors drive incorporation of Tetherin dimers into virion envelopes [14] , [19] . The primary sequence of the Tetherin protein is relatively unimportant for activity , leading to a model in which Tetherin directly tethers virions to infected cells simply through the partition of its membrane anchors into both virion and cell membranes [14] . Consistent with this notion , Tetherin colocalizes with virions at the cell surface and can be observed to reside between cell and tethered virion membranes by electron microscopy [14] , [19] , [20] . Human immunodeficiency virus type-1 ( HIV-1 ) can overcome the restriction imposed by Tetherin through the activity of viral protein U ( Vpu ) [1] , [2] . Vpu is a single-pass type-I membrane protein containing an amino-terminal TMD and a carboxy-terminal cytoplasmic tail domain ( CTD ) . The Vpu CTD consists of two α-helices ( H1 and H2 ) , linked by a short loop [21] , [22] . The loop contains two phosphorylatable serine residues , S52 and S56 [23] , [24] , that are required to recruit β-TrCP and its associated Skp1/Cullin/F-Box ( SCF ) ubiquitin ligase [25] . Vpu can mediate the surface downregulation [2] , [26]–[32] and degradation [33]–[36] of Tetherin , and the phosphorylatable serine residues that are necessary for β-TrCP recruitment are also required for the surface downregulation [2] , [28]–[31] , [33]–[36] and degradation [28] , [31] , [33]–[36] of Tetherin . Analysis of sequences from various mammalian species has revealed that Tetherin has evolved under positive or diversifying selection [37] , [38] and that , consequently , the ability of Vpu to antagonize Tetherin is host-species specific [37]–[40] . Experiments in which domains of human Tetherin were exchanged with those of Tetherin proteins from other primates , revealed that the TMD of Tetherin governs Vpu sensitivity [37]–[40] . Indeed , mutations at two positions in the TMD of human Tetherin , which mimic sequences found in rhesus macaques and African green monkeys , conferred resistance to Vpu antagonism [38] . A few studies have shown that Vpu and Tetherin can be coimmunoprecipitated in a manner that is dependent on Tetherin and Vpu TMDs sequences [34] , [40]–[44] . These studies suggest , but do not prove , that Vpu and Tetherin may interact via their TMDs . In this study , we use cysteine-scanning mutagenesis combined with oxidative cross-linking [45] to: ( i ) demonstrate conclusively that Vpu and Tetherin indeed bind directly to each other in living cell membranes via sequences in their respective TMDs and ( ii ) identify TMD amino acids that are in close proximity to each other in Vpu-Tetherin complexes . While the binding of Vpu to Tetherin appears to be necessary for antagonism , it is clearly not sufficient . Indeed , several studies have reported that Vpu antagonizes Tetherin by causing CTD-dependent sequestration into internal compartments [26]–[32] and degradation via β-TrCP induced ubiquitination [31] , [33]–[36] , [46] . Nonetheless , Vpu apparently induces Tetherin downregulation and degradation in only a subset of cell types , even though all cells tested support Vpu activity [28] , [47] , [48] . Additionally , a Vpu ( Vpu2/6 ) mutant , in which the two phosphorylated serine residues are mutated , and therefore cannot recruit β-TrCP [25] , retains some ability to antagonize Tetherin without inducing cell surface downregulation or degradation [28] , [33] , [34] , [49] . These data suggest that Vpu can inhibit Tetherin function independently of downregulation and degradation . Here , we demonstrate that this is indeed the case , and we show that the Vpu CTD has an autonomous activity , conferred largely by the H2 domain that causes displacement of the Tetherin protein from assembling virions . Moreover , we find that an ExxxLV motif within H2 that has previously been reported to be required for downregulation and degradation of Tetherin [50] is also required for the efficient displacement of Tetherin from sites of virion assembly at the cell surface .
We and others have demonstrated that the HIV-1 Vpu protein selectively antagonizes Tetherin proteins from particular species , in part because Tetherin has evolved under positive selection [37] , [38] . As part of this work , we derived the human TetherinΔGI-T45I mutant that is resistant to antagonism by HIV-1 Vpu due to mutations at two positions in the TMD ( a T45I substitution and a G27/I28 deletion ) [38] . Consistent with previous observations [34] , [40]–[44] , WT Tetherin-HA was efficiently coprecipitated with Vpu2/6-FLAG , whereas minimal amounts of TetherinΔGI-T45I-HA were coprecipitated ( Figure 1A ) . These data suggest that Tetherin and Vpu might physically interact , and that their TMDs are determinants of such an interaction . In order to determine whether Vpu and Tetherin bind directly to each other , in their natural environment ( i . e . the membranes of living cells ) , we attempted to generate chemical crosslinks between the two proteins under native conditions . Specifically , panels of Vpu and Tetherin TMD mutants were generated in which each mutant protein harbored a single cysteine residue in its TMD domain . If Tetherin and Vpu directly bind to each other , a disulfide bond should form between proximal cysteines if the two proteins are coexpressed in an oxidizing environment [45] . To simplify this analysis , we used Tetherin proteins in which the two naturally occurring cysteines in the cytoplasmic domain were mutated to serine ( TetherinCC/SS-HA ) . This did not affect Tetherin activity or sensitivity to antagonism by Vpu ( unpublished observations ) , but removed a potential source of confounding crosslinks . We also used a mutant Vpu2/6-FLAG protein , so that any complexes that might form would not be degraded . Initially , we attempted to generate cysteine cross-links at residues proximal to Tetherin T45 to assess whether this functionally important residue could directly contact Vpu . Thus , Tetherin-HA proteins bearing cysteine residues at positions T45 and I46 were coexpressed with Vpu2/6-FLAG proteins encoding cysteines at positions 2 through 7 . After oxidation , cell lysis , denaturation and immunoprecipitation , proteins were separated on non-reducing SDS-PAGE gels and subjected to Western blot analyses . As expected , the bulk of the Tetherin-HA protein migrated as a homodimer of ∼60 kDa ( Figure 1B ) and the majority of the Vpu2/6-FLAG proteins migrated as ∼17 kDa monomers or ∼35 kDa dimers when whole cell lysates were analyzed . However , in some of the anti-HA immunoprecipitates , Vpu2/6-FLAG was observed to migrate as a shifted ∼90–100 kDa species , consistent with the notion that one or two Vpu2/6-FLAG proteins were crosslinked to a Tetherin dimer ( Figure 1B , lower panel ) . We observed cross-linked Tetherin-Vpu species only when particular residues in the TMD of two proteins were mutated to a cysteine . Specifically , Tetherin proteins bearing Cys residues at TM positions 45 and 46 formed cross-linked species with Vpu proteins modified with cysteines at positions 3 and 4 , but not at positions 2 , 5 , 6 or 7 ( Figure 1B ) . To further confirm that the high molecular weight species were specifically crosslinked complexes that occurred as a result of Tetherin-Vpu binding , we performed a competition assay in which increasing levels of un-tagged and non-cross-linkable TetherinCC/SS were coexpressed with constant amounts of cross-linkable TetherinCC/SS- ( I46C ) -HA and Vpu ( I4C ) -FLAG ( Figure 1C ) . In this and subsequent experiments , whole cell lysates rather than immunoprecipitates were analyzed by Western blotting to determine whether cross-linked Vpu-Tetherin species were generated , which could be observed by the formation of a higher molecular weight FLAG-tagged protein species . Importantly , co-expression of the non-cross-linkable Tetherin reduced the levels of cross-linked Vpu-Tetherin-HA high molecular weight complexes migrating at ∼70 kDa ( Figure 1C , upper panels “XL” label ) . Aliquots from these samples were reduced to dissociate the disulfide linked complexes and re-analyzed by Western blot to confirm that approximately equal levels of the tagged Tetherin and Vpu proteins were present under each condition ( Figure 1C , lower panels ) . Note that the gel shown in Figure 1C and subsequent experiments were run using a different buffer system ( NuPage , Invitrogen ) that is less denaturing than that shown in Figure 1B ( Bio-Rad ) and , therefore , multiple higher-order molecular weight forms of Vpu ( monomers , dimers , and trimers ) can be observed under nonreducing conditions and also note that the crosslinked species migrates at an apparently lower molecular weight ( Figure 1C , upper panel's arrows and XL label , respectively ) . As a further test of the specificity of the crosslinking assay , a cross-linkable form of the TetherinΔGI-T45I-HA mutant was generated and co-expressed with Vpu- ( I4C ) -FLAG . This Vpu-resistant Tetherin mutant was also modified to contain a cysteine at position 46 ( I46C ) or at position 43 ( I43C ) . As before , TetherinCC/SS- ( I46C ) -HA was robustly cross-linked with Vpu- ( I4C ) -FLAG , yielding a protein species migrating at ∼70 kDa that was detected using antibodies to FLAG or HA ( Figure 1D , upper panels “XL” label ) . However the abundance of the Tetherin-Vpu crosslinked complex was greatly reduced when the Vpu-resistant TetherinCC/SS-ΔGI-T45I- ( I46C ) -HA was used . Overall , these data demonstrate that Tetherin and Vpu directly bind to each other via their TMDs in cell membranes , and that this interaction is inhibited by mutations in the Tetherin TMD that confer resistance to Vpu antagonism . We applied the cysteine scanning mutagenesis and cross-linking technique to identify residues that participate in the interactions between Vpu and Tetherin . Each Vpu TMD cysteine mutant ( residues 2 through 27 ) was co-expressed with a panel of five individual , consecutive Tetherin TMD cysteine mutants that were predicted to be in the vicinity of each introduced Vpu cysteine . As the Vpu mutation-scan progressed through the TMD , the register of the five Tetherin TMD cysteine mutants ( between residues 22 and 46 ) was moved to match the Vpu progression . Overall , 130 individual Vpu-Tetherin cysteine-mutant pairs were tested for their ability to form disulfide-linked complexes . As a control , to reveal the background of higher molecular weight FLAG tagged Vpu-Vpu homo-oligomers , each cysteine-modified Vpu protein was also coexpressed with the non-cross-linkable TetherinCC/SS-HA protein . Thus , Vpu-Tetherin cross-linked complexes were revealed by the presence of a new ∼70 kDa protein species on Vpu-FLAG blots , that was absent when the non-crosslinkable TetherinCC/SS-HA was expressed ( labeled “WT” with respect to the TMD ) ( Figure 2 ) . These experiments revealed four major points of proximity between Vpu and Tetherin TMDs ( Figure 2 ) . At the amino terminus of Vpu , ( i . e . close to the outer leaflet of the cell membrane ) VpuQ2C , P3C , and I4C formed cross-links with a few Tetherin cysteine residues . However , the most efficient cross-linking was observed between VpuI4C and TetherinI46C or T45C . Indeed , for these pairs of crosslinkable mutants , the majority of both the Vpu and the Tetherin proteins were present in a cross-linked complex ( Figure 2 , labeled as “XL” ) . Moving further into the TMD , VpuA7C cross-linked with moderate efficiency with TetherinL41C and at low efficiency with Tetherin P40C . At the carboxyl end of the Vpu TMD ( near the cytoplasmic leaflet of the cell membrane ) , VpuV20C formed cross-links with TetherinL29C , and VpuV21C formed cross-links , albeit inefficiently , with TetherinL29C , I28C and G27C . At the extreme C-terminus of the Vpu TMD ( and the N-terminus of the Tetherin TMD ) cross-linking events were observed between VpuV25C and TetherinL24C/L23C/L22C , VpuI26C and TetherinL24C/L23C/L22C , as well as VpuI27C and TetherinL22C . These results are summarized in Table 1 , as a cartoon diagram ( Figure 3 ) and a 3D stereoscopic model ( Figure S1 ) based on NMR-based structural analysis of the Tetherin TMD [44] . Notably , sites of interaction between Vpu and Tetherin , determined by cross-linking , were proximal to some sites that have been reported to be important for the ability of Vpu to antagonize Tetherin and for coimmunoprecipitation [40] , [42]–[44] , ( Table 1 ) . However , cross-linking was not observed at amino acid residues that are predicted to be buried deep within the membrane , including at some sites that have been reported to be important for Tetherin antagonism by Vpu , or for coimmunoprecipitation . For example , Vpu residues A10 , A14 , A18 and W22 have been reported to be important for Vpu function [43] , [44] . However , it was possible that mutating these residues to cysteine for the purposes of our crosslinking experiment disrupted the Vpu-Tetherin interaction and therefore inhibited cross-link formation . To test this hypothesis , we introduced cysteine residues at each of these positions in the context of the HIV-1 proviral plasmid and measured virion release in the presence of Tetherin . Notably , proviral plasmids encoding the Vpu mutants A10C , A14C , and A18C were each capable of generating virions as efficiently as a proviral plasmid encoding WT Vpu ( Figure 4A , B ) . Proviral plasmids encoding the Vpu W22C mutant yielded ∼10-fold fewer virions , however , expression of the VpuW22C mutant was greatly diminished as compared to WT Vpu , which could be responsible for its apparently reduced activity against Tetherin ( Figure 4A , B , C ) . We also tested the ability of this panel of Vpu mutants to downregulate and degrade Tetherin . Again , Vpu mutants A10C , A14C and A18C were indistinguishable from WT Vpu , in these assays , while the poorly expressed VpuW22C mutant was partially able to downregulate and degrade Tetherin ( Figure 5A , B , C ) . We conclude that cysteine mutagenesis of the residues that were previously reported to be important for Vpu function did not , in fact , disrupt Vpu function , except in the case of W22 where expression levels were impaired . We also note that studies in which these residues were identified as important employed an approach in which amino acids with bulky side chains were used to replace alanine . Our cysteine based mutagenesis clearly did not abolish Vpu function , and presumably therefore the ability of Vpu to interact with Tetherin . Therefore , we concluded that these residues either do not contact Tetherin directly , or that the oxidation reagent did not completely penetrate the cell membrane , as has previously been surmised [51] . While residues in the center of the membrane embedded helices may have been inaccessible to the crosslinking reagent , these experiments nevertheless revealed four major points of contact between Vpu and Tetherin , centered on Vpu residues I4 , A7 , V20 , and I26 ( Figures 2 and 3 ) . To determine whether these contact points were required for Vpu function , we mutated four residues in Vpu ( I4A , A7L , V20A , and I26A ) generating a mutant termed VpuQuad . This approach was taken because it has often been found that more than one Vpu mutation is necessary to inhibit Tetherin antagonism ( Table 1 ) . First , the VpuQuad mutant was expressed in the context of a HIV-1 proviral plasmid , in the presence of WT Tetherin . Surprisingly , Vpu Quad was expressed at higher levels as compared to WT Vpu but was nevertheless partially defective as compared to WT Vpu in virion release assays , and a provirus encoding the VpuQuad mutant yielded approximately 5-fold fewer infectious virions in the presence of Tetherin ( Figure 6A , B , C ) . Concordantly , the VpuQuad mutant displayed a partial defect in its abilities to downregulate and degrade Tetherin ( Figure 6D , E , F ) . In conclusion , even though this crosslinking approach indicates that there are multiple sites of contact between Vpu and Tetherin TMDs , it likely underestimates the full extent of the interaction , perhaps because of the inability of the oxidizer to penetrate the membrane . Overall it appears that the interaction between Vpu and Tetherin TMDs is extensive . Indeed , even mutation of four amino acids that can be demonstrated to be in close proximity to Tetherin in Vpu-Tetherin complexes resulted in a mutant Vpu ( VpuQuad ) whose activity was only partly attenuated . Nonetheless , these analyses clearly demonstrate that Vpu and Tetherin TMDs bind directly to each other at several points , and in a physiological context ( i . e . a cell membrane ) . The binding of the Vpu TMD to the Tetherin TMD appears necessary for Vpu function , but may serve simply to recruit the cytoplasmic domain of Vpu to the vicinity of Tetherin . Indeed , elements of the Vpu CTD have been shown to be important determinants of Tetherin antagonist activity , including residues between the TMD and H1 [26] , H2 [26] , an ExxxLV motif in H2 [50] , and the two phosphorylatable serines ( mutated in Vpu2/6 ) that are known to be required for β-TrCP recruitment [25] and for Tetherin downregulation [2] , [28]–[31] , [33]–[36] and degradation [28] , [31] , [33]–[36] . Therefore , we next sought to determine how each of the aforementioned elements of the Vpu CTD contribute Tetherin antagonist function . Consistent with some previous observations [28] , [33] , [34] , [49] , we found that the Vpu2/6 mutant retained partial Tetherin antagonist activity ( Figure 7A , B ) , and was expressed at similar levels to WT Vpu ( Figure 7C ) . Indeed , assessment of a panel of Vpu cytoplasmic tail mutants suggested that multiple elements in the Vpu CTD contribute to Tetherin antagonist function ( Figure 7A , B , C ) . Specifically , Vpu mutants bearing deletions of H1 ( VpuΔH1 ) or H2 ( VpuΔH2 ) , and mutants bearing a mutation of the β-TrCP binding site ( Vpu2/6 ) or the H2 ExxxLV motif to AxxxAA ( VpuH2A3 ) all exhibited residual ability to antagonize Tetherin ( Figure 7A ) . Each of these Vpu mutants , expressed in the context of an HIV-1 proviral plasmid , was impaired as compared to WT Vpu , exhibiting about 10% of WT Vpu activity , but each was able to increase the yield of infectious HIV-1 virions by ∼10 fold , as compared to a ΔVpu proviral plasmid ( Figure 7A ) . Some combinations of mutations ( e . g . combining the H1 deletion with the β-TrCP binding site mutation ( in Vpu2/6ΔH1 ) ) resulted in a Vpu protein that retained the same residual activity of each of the individual mutants ( Figure 7A ) . In contrast , combining , for example , the H2 deletion with the β-TrCP binding site mutation ( in Vpu2/6ΔH2 ) resulted in a Vpu protein that was completely inactive ( Figure 7A ) , even though , Vpu2/6ΔH2 and VpuΔH2 were expressed at the same level ( Figure 7C ) . Combining the H2A3 motif mutation with the Vpu2/6 binding mutation ( in Vpu2/6H2A3 ) resulted in a Vpu protein with reduced activity as compared to the VpuH2A3 or the Vpu2/6 mutants . In general , these data suggest that the Vpu CTD contains multiple elements that contribute to overall Tetherin antagonist function . To determine the role of Vpu CTD domains in degradation of Tetherin and/or downregulation of Tetherin from the cell surface , viruses expressing a panel of Vpu mutants were assessed for these two functions in infected cells ( Figure 8A , B , C ) . As expected , WT Vpu induced surface downregulation and degradation of Tetherin ( Figure 8A , B ) . In contrast , all of the Vpu CTD mutants were either devoid of Tetherin surface downregulation/degradation activity , or impaired in their ability to mediate these effects ( Figure 8B , C ) . This suggested that multiple elements within the CTD , or perhaps the overall protein structure of the Vpu CTD are important for Tetherin downregulation and degradation . However , the VpuΔH2 mutant retained some ability to induce Tetherin downregulation and degradation suggesting that the H2 domain is not absolutely required for these activities ( Figure 8B , C ) . Nonetheless the H2 domain and the ExxxLV motif within it clearly contributed to Tetherin antagonist activity ( Figure 7A ) . Notably , some of the Vpu mutants ( for example Vpu2/6 , VpuΔH1 , Vpu2/6ΔH1 ) appeared to completely lack the ability to induce Tetherin surface downregulation and degradation ( Figure 8 ) , yet retained some ability to antagonize Tetherin activity ( Figure 7 ) . To determine how Vpu impairs Tetherin activity in the absence of downregulation and degradation , we examined the distribution of cell surface Tetherin in cells in which fluorescently labeled HIV-1 particles were assembling . As we and others have previously shown [4] , [19] , [52] , Tetherin colocalized with virions at the plasma membrane ( PM ) in the absence of Vpu ( Figure 9A ) . Conversely , Tetherin-virion colocalization was inhibited in the presence of WT Vpu ( Figure 9A ) . The TetherinΔGI-T45I mutant , which exhibits reduced ability to bind Vpu ( Figure 1A and 1D ) , colocalized with virions even in the presence of Vpu , consistent with the notion that displacement of Tetherin from nascent virions at the cell surface is dependent on binding between the Vpu and Tetherin TMDs ( Figure 9B ) . These data are quantified and are summarized in Figure 9C . Notably , the Vpu2/6 mutant , which was devoid of Tetherin downregulation and degradation activity , also specifically inhibited WT Tetherin but not TetherinΔGI-T45I colocalization with virions ( Figure 9A , B , lower row of panels ) . Analysis of a wider panel of Vpu CTD mutants revealed that VpuΔH1 , and Vpu2/6ΔH1 mutants retained full ability to inhibit colocalization of Tetherin with nascent virions . These Tetherin mutants each lacked downregulation and degradation activity , but notably , encoded an intact H2 domain ( Figure 9C , D ) . The ability of Vpu to inhibit colocalization of Tetherin with virions was reduced slightly by the VpuΔH2 mutation ( Figure 9C , D ) , but it is important to note that this mutant is also capable of promoting Tetherin downregulation and degradation ( Figure 8B , C ) . However , when Tetherin downregulation and degradation was also abrogated by inclusion of the Vpu2/6 mutation ( in Vpu2/6ΔH2 ) , this protein was only poorly able to displace Tetherin from sites of viral assembly ( Figure 9C , D ) . The Vpu2/6H2A3 mutant was even less able to displace Tetherin from sites of viral assembly compared to Vpu2/6ΔH2 ( Figure 9D ) . Thus , these data suggest that the Vpu H2 domain , and the ExxxLV motif within it , was important for the displacement of Tetherin from sites of viral assembly . Moreover , the ability of Vpu mutants to displace Tetherin from sites of virion assembly may explain how they are able to partially antagonize Tetherin in the absence of Tetherin downregulation or degradation . While the Vpu TMD is necessary for Tetherin binding , the above data indicate that various additional Vpu CTD sequences are also required for Tetherin antagonism , and the displacement of Tetherin from virions . To determine whether Vpu TMD–Tetherin TMD interactions are necessary for the ability of Vpu to displace Tetherin from nascent virions , or simply provide a means by which Vpu is recruited to Tetherin , we adopted an experimental strategy in which the Vpu CTD was artificially recruited to Tetherin in the absence of the TMD . Specifically , we engineered proteins in which either the entire Vpu CTD , or individual elements thereof , were fused to the N-terminus of Tetherin ( Figure 10A–E ) . An additional advantage of this approach is that it ensures that every Tetherin molecule is associated with a Vpu CTD or CTD fragment at a stoichiometric ratio of 1∶1 . To retain the topological relationship of the Vpu CTD with the cell membrane , we appended its N-terminus with a sequence of amino acids ( MGCGCSSHPEGGG ) from the N-terminus of Lck , generating a palmitoylated and myristoylated membrane anchor ( Figure 10A–D ) [53]–[55] . Transient expression of the modified Tetherin proteins revealed that each was expressed at the cell surface ( Figure S2A ) . Indeed , transiently expressed WT Tetherin , Lck-Tetherin and Lck-H2-Tetherin , were present on at the cell surface at similar levels ( Figure S2A ) . However , both Lck-CTD2/6-Tetherin and especially Lck-H1-Tetherin were expressed at the cell surface at ∼3-fold and ∼5-fold lower levels respectively ( Figure S2A ) . To facilitate microscopic analysis of the localization of the Tetherin proteins we generated clonal cell lines that expressed each of the modified Tetherin proteins at similar levels ( insofar as was possible ) at the cell surface ( Figure 10F ) . In this panel of cell lines , the levels of the Lck-Tetherin proteins were more modest than in transiently transfected cells , and quite well matched , with the exception of the Lck-H1-Tetherin , which only yielded clones expressing lower levels of Tetherin at the cell surface . Microscopic examination of these cell lines revealed that when the Tetherin N-terminus was appended with the Lck N-terminus alone , ( in Lck-Tetherin , Figure 10A ) The modified Lck-Tetherin protein exhibited a subcellular localization that was indistinguishable from the unmodified Tetherin protein ( Figure 10E ) . The Lck-Tetherin protein that was appended with the H2 domain of Vpu CTD ( Lck-H2-Tetherin , Figure 10D ) exhibited a subcellular localization that was similar to WT Tetherin or Lck Tetherin , with prominent cell surface staining . In contrast , the Tetherins appended with the entire Vpu CTD2/6 ( Lck-CTD2/6-Tetherin , Figure 10B ) , or particularly the H1 domain alone ( Lck-H1-Tetherin , Figure 10C ) were less prominently localized to the cell surface . Indeed , the Lck-H1-Tetherin fusion protein appeared to localize in part in a perinuclear ring ( Figure 10C ) and colocalized with a transiently expressed a DsRED-ER marker ( Pearson's coefficient = 0 . 6 ) whereas , the Lck-Tetherin protein exhibited only minimal colocalization with the ER marker ( Pearson's coefficient = 0 . 02 ) ( Figure S2B ) . Consistent with the notion that H1 induces ER retention , western blot analysis of the clonal stable cell lines expressing the Lck-Tetherin proteins revealed that Lck-H1-Tetherin dimers were of lower molecular weight than that the other Tetherin proteins , presumably because they failed to acquire the high molecular weight complex carbohydrates that are characteristic of Tetherin proteins that have proceeded through the secretory pathway [14] . These data are also consistent with previous observations demonstrating that a Vpu protein harboring a C-terminal truncation can modify the subcellular localization of Tetherin [26] . Overall these findings suggest the presence of an ER membrane retention activity in H1 . However , this property was less evident in the Lck-CTD2/6-Tetherin ( Figure 10B ) , suggesting the possibility that H2 suppresses the putative ER retention signal in H1 . Not only did the modified Lck-Tetherin protein ( Figure 10A ) exhibit a subcellular localization that was indistinguishable from the unmodified Tetherin protein ( Figure 10E ) , but it also restricted virion release , and retained Vpu sensitivity ( Figure 11A , B ) . Thus , the Lck modification was , by itself , apparently inert . Analyses of virion release in the presence of the various Lck-Vpu-Tetherin fusion proteins revealed that the entire CTD2/6 , as well as the individual H1 or H2 domains , were able to antagonize the antiviral activity of Tetherin when fused to its N-terminus ( Figure 11A , B ) . The inability of Lck-H1-Tetherin to inhibit HIV-1 particle release might be explained by the redistribution of the modified Tetherin protein away from the PM ( Figures 10 , Figure S2 ) . However , Lck-H2-Tetherin was abundantly expressed at the PM ( Figure 10 , Figure S2 ) , suggesting that Vpu H2 harbors an activity that antagonizes Tetherin , without removing it from the PM ( Figure 11A , B ) . The apparent ability of H2 to antagonize Tetherin in the context of the Lck-Tetherin fusion protein was abolished when ExxxLV motif in H2 was mutated to AxxxAA in Lck-H2A3-Tetherin , ( Figures 11A , 11B , and S2 ) . Thus , these data demonstrate that the Vpu H2 domain and ExxxLV motif within it contribute to Vpu function , in the absence of Tetherin downregulation and degradation . To determine whether the isolated Vpu CTD and fragments thereof could modulate the colocalization of Tetherin with nascent virions , we used fluorescence microscopy to quantify the colocalization between the Lck-Vpu-Tetherin fusion proteins and fluorescently labeled HIV-1 particles . As compared to Lck-Tetherin , the Lck-CTD2/6-Tetherin fusion protein was impaired in its ability to colocalize with virions , ( Figure 12A , B ) . Moreover , Lck-Tetherin constructs encoding either H1 or H2 fragments alone ( Lck-H1-Tetherin and Lck-H2-Tetherin ) appeared to be nearly completely excluded from virions ( Figure 12A , B ) . Note again , the Lck-H1-Tetherin protein was primarily localized to internal compartments , and comparatively poorly expressed on the cell surface , which may be partly responsible for the exclusion of Lck-H1-Tetherin from nascent virions . Conversely , Lck-H2-Tetherin is abundant on the cell surface , but did not colocalize with nascent virions ( Figure 10 , Figure S2 , Figure 12A , B ) . Notably , mutating the ExxxLV motif within H2 abolished the ability of H2 to cause exclusion of Lck-Tetherin from nascent virions ( Figure 12A , B ) . Thus , these experiments demonstrate that the isolated Vpu H2 domain is sufficient to displace Tetherin from sites of viral assembly , in the context of Lck-H2-Tetherin , even when the fusion protein is abundantly expressed at the cell surface . Interestingly , the isolated H1 and H2 domains were slightly more potent than the complete CTD2/6 in inhibiting Lck-Tetherin colocalization with virions , but both were clearly and independently able to inhibit Tetherin colocalization with virions and inhibit antiviral activity ( Figures 11 and 12 ) . To further investigate this apparent displacement activity , we assessed the ability of the Vpu CTD fragments to block incorporation of Tetherin into released virion particles . This strategy was based on our previous finding that removal of the GPI anchor from Tetherin ( TetherinΔGPI ) causes a loss of antiviral activity and efficient incorporation of the TetherinΔGPI protein into released virions [14] . In this experiment , we stably expressed Lck-Tetherin , Lck-TetherinΔGPI , Lck-H2-TetherinΔGPI , or Lck-H1-TetherinΔGPI and infected cells with viruses that encoded Vpu2/6 or lacked Vpu ( ΔVpu ) . All cells that expressed Vpu2/6 efficiently released virions and did not efficiently incorporate Lck-TetherinΔGPI protein into virions ( Figure 13 ) . As before , Lck-Tetherin inhibited virus release in the absence of Vpu ( Figure 11A , B , Figure 13 ) . Conversely , cells expressing Lck-TetherinΔGPI generated similar amounts of released virions as did unmodified cells . Importantly , Lck-TetherinΔGPI was efficiently incorporated into virions , specifically in the absence of Vpu ( Figure 13 ) . Further , both the Vpu H1 and H2 fragments directly appended to Tetherin ( in the context of a Lck-TetherinΔGPI protein ) could inhibit Lck-TetherinΔGPI incorporation into released virions ( Figure 13 ) . As before , Tetherin localization to internal compartments was induced by H1 , likely contributed to the lack of incorporation of Lck-H1-TetherinΔGPI into virions , while Lck-H2-TetherinΔGPI was well expressed at the cell surface ( Figure S3 ) . Therefore , H2 , when directly appended to TetherinΔGPI , inhibited the incorporation of Tetherin into virions without affecting levels of cell surface expression .
Overall the data in this manuscript lead to a model for Vpu function ( Figure 14 ) in which the TMD domain is responsible for directly binding Vpu to Tetherin , while H1 and the phosphorylated serine residues in the CTD are largely responsible for intracellular retention and degradation , and H2 is primarily responsible for displacing Tetherin from virions at the cell surface . Previous studies have demonstrated that the TMD of Tetherin determines its sensitivity to Vpu [37] , [38] , [40] , [42] , [44] and that residues in the Vpu TMD are important determinants of its activity [43] , [44] . Moreover , Vpu and Tetherin can be coimmunoprecipitated and give a positive signal in bimolecular fluorescence complementation assays [34] , [40]–[44] , suggesting that they are components of the same protein complex , and NMR studies indicate that residues in the Tetherin TMD undergo chemical shifts in the presence of a fragment of the Vpu TMD in artificial lipid bilayers [44] . However , no previous study had unequivocally determined whether intact Vpu and Tetherin proteins directly bind to each other in the membranes of living cells . We employed cysteine-scanning mutagenesis coupled with oxidation-induced cross-linking [45] , which enabled us to demonstrate that Vpu and Tetherin interact directly in cell membranes . Additionally , we were able to identify several Tetherin and Vpu residues at the periphery of their respective TMDs that are in close proximity to each other . Previously reports have identified residues within the central portion of the TMD of Vpu that are important for Vpu function , when mutated in combination , and appear to interact with Tetherin [43] , [44] , While we did not observe cross-linking at these positions , nor did we observed functional consequences when these residues were mutated to cysteine , it is possible that the cell membrane inhibits penetration of oxidation reagent required for disulfide bond formation [51] . Thus , while our analysis identifies several points of contact between Tetherin and Vpu , it may underestimate the true extent of the interactions between the Tetherin and Vpu TMDs . Our conclusion that there are multiple points of contact between Tetherin and Vpu TMDs is consistent with previous functional/genetic analyses . For example , mutation of individual residues that exhibit evidence of positive selection in the TMD of human Tetherin to rhesus macaque Tetherin counterparts , resulted in proteins that were at least partially sensitive to Vpu , whereas combinations of mutations gave fully Vpu resistant proteins , as summarized in Table 1 [37] , [38] . Indeed , a variety of published mutational analyses of Tetherin [37] , [38] , [40] , [42] , [44] and Vpu [43] , [44] have revealed that either large or multiple changes are required to completely disrupt the functional interaction between Vpu and Tetherin , suggesting multiple contact points exist between these two proteins , perhaps across almost the entire length of the TMDs ( Table 1 ) . Moreover , the finding that crosslinked residues in the Tetherin TMD appeared to occur on opposing sides of the TMD helix ( Figure S1 ) suggests the possibility that there more an one binding site for Vpu on Tethe , perhaps in the context of a tetherin dimer . Although there is significant concordance between the results of our crosslinking analysis and prior functional genetic analyses ( Table 1 ) , the above considerations likely underlie some of the discrepancies . Our analyses identified Tetherin TMD residues L22 , L23 , L24 , G27 , L29 , L41 , I43 , F44 , T45 , and I46 , as points of proximity to Vpu , and Vpu residues Q2 , P3 , I4 , A7 , V20 , V25 , I26 , I27 as points of proximity to Tetherin , as summarized in Figures 3 and S1 and Table 1 . Previous studies identified Tetherin TMD residues ( L23 , L24 , G27 , G28 , V30 , I33 , I34 , I36 , L37 , P40 , L41 , T45 ) as determinants of Vpu sensitivity and Vpu TMD residues ( A10 , A14 , I15 , A18 , and W22 ) as key determinants of Tetherin antagonist activity ( summarized in Table 1 ) . Additionally , Tetherin residues ( L23 , L24 , I26 , G27 , I28 , V30 , I34 , I37 , and L41 ) and Vpu residues ( A10 , A14 , A18 , and W22 ) have been implicated as potential points of interaction using coimmunoprecipitation or bimolecular fluorescence complementation assays [40] , [42]–[44] . Notable discrepancies between our crosslinking data and prior functional/genetic assays include the observation that no cross-linking involving Tetherin V30 was detected , yet this residue appears to be a determinant of Vpu sensitivity and is involved in a NMR chemical shift [37] , [38] , [44] . We did however detect crosslink involving a neighboring Tetherin residue ( L29 ) suggesting the possibility that perturbations of Tetherin TMD structure induced by V30 mutation might underlie its effect on Vpu sensitivity . Similarly , Vpu W22 , mutation of which impairs Vpu function [43] , [44] , did not appear to directly contact Tetherin , but neighboring residues ( V20 and V21 ) did appear to be in close proximity to Tetherin . Again these data suggest that residues which , when mutated , result in loss of function , may be involved in maintaining the overall structural conformation of TMDs rather than constituting points of interaction . Four residues within Vpu that formed the strongest crosslinks with Tetherin contributed to Vpu function , as the VpuQuad mutant was somewhat attenuated in its ability to antagonize Tetherin . However , these were not the only residues that were found to be in close proximity to Tetherin , and our findings underscore the notion that Vpu and Tetherin interact at multiple points within their respective TMDs as well as the importance of information obtained using different assays in interpreting how proteins interact , and the contribution of each component of the interaction to overall protein activity . Clearly , TMD interactions recruit Vpu and Tetherin to each other , and thereby enable antagonism to occur through activities elicited by the Vpu CTD . Our functional dissection of the Vpu CTD suggested that Vpu possesses a degradation and downregulation-independent mechanism of antagonizing Tetherin . This explains the ability of mutated Vpu proteins to partially abrogate Tetherin antiviral activity in the absence of downregulation or degradation . Specifically , we found that Vpu can cause displacement of Tetherin from assembling virions at the cell surface . Analysis of individual domains within the CTD of Vpu , in the context of mutant Vpu proteins , or when artificially ‘recruited’ to Tetherin in the context of Lck-CTD-Tetherin fusion proteins , revealed that either H1 or H2 domains of the CTD could elicit anti-Tetherin activity . Our analysis complements previous data that determined that an ExxxLV motif within H2 plays a role in directing Tetherin localization [50] . However our data further suggests that the Vpu H2 and the ExxxLV motif within it can act at the PM to displace Tetherin from sites of viral assembly . Precisely what role the ExxxLV motif plays in this process is unclear . It is possible that it is required for the structural integrity of H2 , or forms part of a cofactor binding site . The fact that ExxxLV is required both for displacement of Tetherin from particle assembly sites at the cell surface , as well as determining the intracellular fate of Tetherin [50] suggests that its integrity is generally required for Vpu functions . Conceivably , displacement from assembly sites could be a first step in an H2-dependent multi-step process that leads to the downregulation and degradation of Tetherin . Further analyses revealed that the Lck-H1-Tetherin fusion protein was predominantly localized to the ER . This finding is consistent with previous reports that a Vpu mutant lacking H2 localized to new internal compartments [26] and suggests that Vpu H1 has a discrete ER retention or retrograde trafficking signal , which may play a role in antagonizing Tetherin [56] . Ultimately , in order to be efficiently released by infected cells , assembling virions require Tetherin to be absent from sites of viral assembly . Previous reports have revealed how Vpu causes Tetherin sequestration to internal compartments and downregulation from the cell surface , as well as decreased Tetherin stability [26]–[31] , [33]–[36] , [41] , [49] . Here we have shown that Vpu can also displace Tetherin from sites of virus assembly without removing it from the cell surface , employing an activity that appears largely contained within the H2 domain of the Vpu CTD . Thus , it appears that Vpu can mobilize multiple mechanisms to abrogate the restriction imposed on HIV-1 by Tetherin .
Codon optimized Vpu [57] was mutated to Vpu2/6 ( S52A S56A ) using overlap-extension PCR and then inserted into the expression vector pCR3 . 1 ( Invitrogen ) with the sequence EcoRI-CoVpu-XhoI-FLAG ( MDYKDHG-stop ) -NotI . This Vpu2/6-FLAG expression vector was used as the parental plasmid to mutate each individual TMD amino acid residue to a cysteine residue using mutagenic 5′ PCR primers , or overlap-extension PCR . Construction of plasmids expressing Tetherin-HA and TetherinΔGI-T45I-HA proteins were described previously . Briefly , an HA sequence was inserted at nucleotide position 463 in the Tetherin cDNA . Thereafter , G25 and I26 were deleted and a T45I mutation introduced [38] . WT and T45I-mutant Tetherin cDNAs were then mutated , to generate the parental backbones TetherinCC/SS , harboring C9S and C20S mutations , which were then used to generate the individual cysteine point mutants in the Tetherin TMD . All mutagenesis was accomplished by using overlap-extension PCR . Plasmids expressing Lck-Tetherin-HA , Lck-VpuCTD2/6-Tetherin-HA , Lck-VpuH1-Tetherin-HA , Lck-VpuH2-Tetherin-HA chimeras were generated in the context of a Tetherin-HA expression plasmid using overlap-extension PCR [58] . The cDNAs encoding Tetherins without their GPI anchor ( ΔGPI ) were generated by inserting a stop codon after the inserted HA epitope . The VpuCTD2/6 , H1 and H2 sequences were defined , respectively , as residues E28-L81 , I32-A49 , and E57-L81 of NL4-3 Vpu . The Lck-Tetherin constructs were inserted into the retroviral vector pLHCX ( Clontech ) to generate stable cell lines expressing these proteins . The proviral plasmids pNL4-3 ( WT ) , Vpu deficient pNL4-3 ( ΔVpu ) and pNL4-3 ( Vpu2/6 ) ( S52A , S56A ) were previously described or were generated by overlap-extension PCR [59] . The proviral plasmid pNL4-3 ( MA-YFP ) that contained a YFP cDNA inserted into the stalk region of MA was previously described [60] . The various Vpu mutants , VpuDelR34-E47 ( VpuΔH1 ) , VpuE59Stop ( VpuΔH2 ) , and Vpu2/6ΔH1 were generated using overlap-extension PCR and inserted into pNL4-3 and pNL4-3 MA-YFP via EcoRI/NheI digestion , and into the HIV-1 based vector , V1B-GFP , via MfeI/KpnI digestion and ligation . For some Vpu mutations ( S56A , E59Stop , E59A , L63A , V64A ) changes also introduced mutations into the extreme N-terminus of the overlapping Env ORF . However , these were all at positions that were intrinsically variable in natural HIV-1 sequence , and where possible the Env changes were conservative . None of the Env mutations affected the yield of infectious HIV-1 particels in the absence of Tetherin . All constructs were sequence verified . Oligonucleotide sequences are available upon request . The cells HEK293T and HeLa-TZM cells expressing CD4/CCR5 and a LacZ reporter gene under control of the HIV-1 LTR were maintained in DMEM media containing 10% fetal calf serum and gentamycin ( 2 µg/ml , Gibco ) . HEK293T cells were transduced with pLHCX ( Clontech ) based retroviral vectors expressing genes of interest and selected with hygromycin ( 50 µg/ml ) ( MediaTech , Inc ) to generate cell lines expressing Tetherin-HA , [38] , Lck-Tetherin-HA , Lck-VpuCTD2/6-Tetherin-HA , Lck-VpuH1-Tetherin-HA , Lck-VpuH2-Tetherin-HA , Lck-TetherinΔGPI-HA , Lck-H1-TetherinΔGPI-HA , and Lck-H2-TetherinΔGPI-HA . Oxidation induced cytsteine cross-linking was executed using a protocol adapted from [45] . Specifically , HEK293T cells were seeded at 2 . 5×105 cells/well in 12 well plate and co-transfected the following day with 500 ng of a plasmid expressing a Vpu2/6-FLAG TMD cysteine mutant and 400 ng of a plasmid expressing a TetherinCC/SS-HA TMD cysteine mutant . At 40 hours post transfection , the cells were incubated on ice and the media was replaced with oxidizing solutions . The oxidizing solutions were added to the cells by simultaneously adding 135 µl of each of two TMKP buffer solutions ( 10 mM Tris-HCl ( pH 7 . 5 ) , 15 mM MgCl2 , 10 mM KCL , protease inhibitor cocktail ( 1 tablet/10 ml , Roche ) containing either Copper sulfate ( 4 mM ) or 1 , 10-Phenanthroline ( 12 mM , Sigma ) to give a final concentrations of 2 mM and 6 mM , respectively . After mixing and incubation on ice for ten minutes , the cells were disrupted using a 550 Sonic Dismembrator probe sonicator ( Fisher-Scientific ) at a setting of 2 and incubated on ice for a further 10 mins . The lysates then received 40 µl each of additional TMKP Copper sulfate ( 6 mM ) and Phenanthroline ( 18 mM ) for a final concentration of 2 . 2 and 6 . 6 mM , respectively . The reaction was mixed and incubated for 10 mins on ice and then stopped by the addition of 20 mM N-ethylmaleimide and 5 mM EDTA . Lastly , 4× Nu-PAGE ( Invitrogen ) sample buffer was added ( without reducing agent ) and the samples were analyzed by SDS-PAGE and Western blotting . For spontaneous crosslinking assays in the absence of oxidizing agent , 200 ng of a plasmid expressing a Vpu2/6-FLAG TMD cysteine mutant and 200 ng of TethCC/SS-I46C-HA DNA , and 50 or 100 ng TetherinCC/SS DNA were used to transfect HEK293T cells . At 48 hrs . post-transfection , the cells were harvested with NuPAGE sample buffer without reducing agent . For immunoprecipitation of Vpu-Tetherin cross-linked complexes , HEK293T cells were cotransfected and treated as above with oxidizing agents . Immunoprecipitation conditions were as described previously [61] . After oxidation , cells were lysed in detergent-rich denaturing buffer including protease inhibitor tablets ( Roche ) and 5 mM N-ethylmaleimide to inhibit further cross-linking . After centrifugation to clear cell debris , cell lysates were diluted 5-fold with the same buffer , except SDS was substituted with NP-40 ( to a final concentration of 1% ) , in order to dilute SDS to 0 . 1% prior to immunoprecipitation . The cleared lysates were incubated with anti-HA . 11 monoclonal antibody ( Covance ) , and complexes were isolated with Protein-G Sepharose 4 Fast Flow ( GE Healthcare ) . Immunoprecipitates and unfractionated cell lysates were analyzed by Western blotting . For non-denaturing immunoprecipitation of non-cross-linked Tetherin-Vpu complexes procedures were as described previously [34] . HEK293T cells were seeded in 24 well plates at a concentration of 1 . 5×105 cells/well and transfected the following day using polyethylenimine ( PEI ) ( PolySciences ) with 500 ng of pNL4-3 ( WT ) or pNL4-3 ( ΔVpu ) along with various amount of a Tetherin expression plasmid ( 25 ng to 200 ng ) , with the empty pCR3 . 1 vector used as a filler . Following transfection , media was replaced with fresh DMEM ( 1 ml ) and supernatants were collected and filtered through a 0 . 2 µm PVDF filter ( Millipore ) at 40 hours post transfection . Infectious virion yield was determined by inoculating HeLa-TZM indicator cells seeded in a 96 well plate at 1 . 5×104 cells/well with 10 µl of serially diluted virus . At 48 hrs . post-infection , beta-galactosidase activity was determined using GalactoStar reagent following manufacturer's instructions ( Applied Biosystems ) . Physical particle yield was determined by layering 750 µl of the virion containing supernatant onto 800 µl of 20% sucrose in PBS and centrifugation at 20 , 000×g for 90 minutes at 4°C . Virion pellets were then analyzed by Western blotting . HEK293T cells seeded at 4×106 cells/10 cm plate were cotransfected with VSV-G ( 1 µg ) , pCRV1/HIV-1GagPol ( 5 µg ) , and a V1B-GFP based HIV-1 vector expressing a WT or mutant Vpu protein DNA ( 5 µg ) . At 40 hours post-transfection , the supernatants were filtered ( 0 . 2 µm PVDF , Millipore ) . The virus titers were determined by measuring the fraction of GFP positive cells after infection of HEK293T cells with serially diluted virus . Thereafter , HEK293T cells stably expressing Tetherin-HA were seeded at 1 . 5×105 cells/well in a 24 well plate . The following day , cells were transduced at a MOI of 0 . 3 with the aforementioned V1B-GFP vectors expressing various Vpu proteins . After 48 hrs , the cells were harvested with 5 mM EDTA in PBS and blocked with 3% rabbit serum in PBS at 12°C and incubated with rabbit anti-HA conjugated to SureLight APC 1∶50 ( Colombia Biosciences ) or anti-huTetherin antibody conjugated to APC ( Biolegend ) and DAPI ( Invitrogen ) . FACS analysis using a FACS ( BD LSR II ) was used to determine the relative amount of surface Tetherin expression in GFP-positive and DAPI-negative cells . Alternatively , cells were infected with the same panel of viruses at an MOI of 3 , lysed 40 hours later with NuPage sample buffer , and Tetherin expression analyzed by Western blotting . Pelleted virions , cell lysates and immunoprecipitates were resuspended in SDS-PAGE loading buffer and separated on NuPAGE Novex 4–12% Bis-Tris Mini Gels ( Invitrogen ) or on BioRad 12% Tris-HCl gels with Laemmli sample buffer . Proteins were blotted onto nitrocellulose membranes ( HyBond , GE-Healthcare ) . The blots were then probed with mouse anti-HIV-1 capsid ( NIH ) , rabbit anti-HA ( Rockland ) , mouse anti-FLAG ( Sigma ) , mouse anti-GFP ( Roche ) or rabbit anti-tubulin ( Santa-Cruz Biotechnology ) primary antibodies . Bound primary antibodies were detected using HRP-conjugated secondary antibodies ( Jackson Laboratories ) and chemiluminescent detection reagents ( Pierce ) or with fluorescently labeled secondary antibodies , and a LI-COR Odyssey scanner ( LI-COR biosciences ) . To determine the subcellular distribution of Tetherin mutants and chimeras , HEK293T cell lines stably expressing HA-tagged Tetherin , Lck-Tetherin , Lck-VpuCTD2/6-Tetherin , Lck-VpuH1-Tetherin , or Lck-VpuH2-Tetherin were seeded at 1×105 cells on 3 . 5-cm glass bottomed dishes coated with poly-L-lysine ( Mattek ) . Two days later the cells were fixed with 4% paraformaldehyde ( PFA ) , washed with 5 mM glycine in PBS followed by 0 . 5% TritonX-100 to permeabilize the cells . Cells were blocked with 1% bovine serum albumin in PBS and incubated with anti-HA . 11 monoclonal antibody ( Covance ) followed by goat antimouse IgG conjugated to Alexafluor-594 ( Molecular Probes ) . Cells were counterstained with DAPI ( Invitrogen ) . To assess the colocalization of Tetherin with virus particles , cell lines expressing Tetherin mutants or chimeras were seeded at 5×104 cells per/dish . At 48 hrs later , the cells were cotransfected , using PEI , with pNL4-3 ( WT ) , pNL4-3 ( ΔVpu ) . pNL4-3 ( Vpu2/6 ) pNL4 . 3 ( VpuΔH1 ) , pNL4 . 3 ( Vpu2/6ΔH1 ) , pNL4 . 3 ( VpuΔH2 ) , pNL4 . 3 ( Vpu2/6ΔH2 ) , or pNL4 . 3 ( Vpu2/6H2A3 ) , mixed with a matching counterpart expressing MA-YFP at a ratio of 1∶1 ( 200 ng each ) . At 48 hours post-transfection , cells were processed as described above , but were not permeabilized . Imaging was performed as described previously [62] , but in brief , a z-series of images were taken to capture the top of the cell . Several regions of each dorsal cell surface were selected to measure colocalization between Tetherin and viral particles and the mean Pearson's correlation coefficient calculated using DeltaVision software , as previously described [62] . To assess the incorporation of Tetherin into virions a protocol was adapted from previous work [14] . HEK293T cell lines stably expressing Lck-Tetherin , Lck-TetherinΔGPI , Lck-H2-TetherinΔGPI , or Lck-H1-TetherinΔGPI were seeded at a concentration of 5×105 cells/well in 6 well plates and infected the next day by VSV-G pseudotyped NL4-3 HIV-1 ( Vpu2/6 ) ΔNef::GFP or NL4-3 HIV-1 ( ΔVpu ) ΔNef::GFP virus at an MOI of 1 . The next day the media was replaced with 4 mls of fresh media . The supernatants were harvested the following day , filtered ( 0 . 2 µm PVDF , Millipore ) , layered onto 8 mls of 20% sucrose in PBS and centrifuged at 90 , 000×g for 90 minutes . The isolated viral pellets were suspended in NuPAGE sample loading buffer and the whole cell lysates were analyzed by Western blotting . | At the cell surface , HIV-1 particles are assembled and then released to infect new cells . However , an anti-viral host restriction factor , Tetherin , can tether outgoing virions to the infected cell surface , preventing their dissemination . HIV-1 overcomes this block through the expression of the viral accessory protein Vpu , which antagonizes Tetherin . In this study , we demonstrate that the domains of Vpu and Tetherin that are embedded in the outer cell membrane bind directly to each other within the membrane , and we identify amino acids that participate directly in the interaction between these two proteins . After binding to Tetherin , Vpu can induce its removal from the cell surface and degradation . However , a mutant Vpu lacking these activities retains some capacity to antagonize Tetherin . We show that this residual activity requires a particular portion of the intracellular domain of Vpu and is manifested as an ability to displace Tetherin from sites of viral assembly , without affecting the overall level of Tetherin at the cell surface . These data indicate that Vpu directly binds to Tetherin and then employs multiple mechanisms , including displacement , to counteract Tetherin's ability to restrict virus particle release . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology",
"biology",
"microbiology"
] | 2013 | Vpu Binds Directly to Tetherin and Displaces It from Nascent Virions |
Multicellular differentiated organisms are composed of cells that begin by developing from a single pluripotent germ cell . In many organisms , a proportion of cells differentiate into specialized somatic cells . Whether these cells lose their pluripotency or are able to reverse their differentiated state has important consequences . Reversibly differentiated cells can potentially regenerate parts of an organism and allow reproduction through fragmentation . In many organisms , however , somatic differentiation is terminal , thereby restricting the developmental paths to reproduction . The reason why terminal differentiation is a common developmental strategy remains unexplored . To understand the conditions that affect the evolution of terminal versus reversible differentiation , we developed a computational model inspired by differentiating cyanobacteria . We simulated the evolution of a population of two cell types –nitrogen fixing or photosynthetic– that exchange resources . The traits that control differentiation rates between cell types are allowed to evolve in the model . Although the topology of cell interactions and differentiation costs play a role in the evolution of terminal and reversible differentiation , the most important factor is the difference in division rates between cell types . Faster dividing cells always evolve to become the germ line . Our results explain why most multicellular differentiated cyanobacteria have terminally differentiated cells , while some have reversibly differentiated cells . We further observed that symbioses involving two cooperating lineages can evolve under conditions where aggregate size , connectivity , and differentiation costs are high . This may explain why plants engage in symbiotic interactions with diazotrophic bacteria .
The reproduction and development of differentiated multicellular organisms follows a complex iterative pattern . Almost all differentiated multicellular organisms develop from a single pluripotent germ cell that divides and differentiates . Although terminally differentiated somatic cells contain all the necessary genetic information to produce whole organisms [1]–[3] , they are unable to do so despite the potential cost in reproductive opportunities for the organism . In contrast , organisms composed of reversibly differentiated cells can reproduce through fragmentation or budding . Examples include most plants , and some animals such as corals , hydra , planarians , several echinoderms , and some annelid worms [4]–[7] . In these organisms , a fragment can regenerate the missing parts of the organism , resulting in several complete new organisms . During such regeneration , somatic cells in the fragments can sometimes de-differentiate and form a blastema ( a group of undifferentiated cells ) that regenerates the missing parts [6] . This means that somatic cells undergo reversible differentiation , and can revert back to their undifferentiated forms . Multicellular cyanobacteria are some of the simplest multicellular organisms known . They are of particular interest because in some species , cells are terminally differentiated [8] , [9] , while in others , terminally differentiated cells have not been observed . Cyanobacteria have very diverse morphologies . They are found as single cells , multicellular filaments of undifferentiated cells , and differentiated multicellular filaments ( with or without branching ) [10] . In differentiated multicellular cyanobacteria , some cells specialise in photosynthesis while others specialise in nitrogen fixation . Only one genus of cyanobacteria ( Trichodesmium ) is known that could potentially exhibit reversible differentiation [11] , [12] . In contrast , several terminally differentiating cyanobacteria are known , of which two examples are the genera Anabaena and Nostoc . These cyanobacteria are composed of two cell types: the vegetative cell ( germline ) and the heterocyst cell ( soma ) . Vegetative cells are photosynthetic , reproduce through division , and are able to differentiate into heterocyst cells [13] . Heterocysts do not divide , have a thicker cell wall , and perform nitrogen fixation . They are also larger than vegetative cells . In this manner , vegetative cells obtain fixed nitrogen from heterocysts , and heterocysts obtain fixed carbon from the vegetative cells . These cyanobacteria have strongly regulated patterns of differentiation , forming heterocysts every 11 vegetative cells , with little variance in the number of vegetative cells between heterocysts [14] . Since the pattern of differentiation of cyanobacteria can not be explained solely through random differentiation [13] , many studies have focused on understanding the mechanisms of pattern formation [13] , [15]–[17] . Experimental evidence has identified three genes that play a key role in its regulation . NtcA , HetR and PatS all play a role in the differentiation mechanism of cyanobacteria . NtcA is a DNA binding factor that regulates the transcription of genes involved in nitrate transport and assimilation [18] , HetR has been shown to be expressed shortly after heterocyst formation is induced when fixed nitrogen becomes scarce [19] . PatS is a gene that represses the formation of heterocysts and is believed to be produced by developing heterocysts and released to neighboring cells to prevent the formation of clusters of tightly spaced heterocysts [14] . A proximal explanation for the fact that heterocysts are terminally differentiated may be the physical constraints on cell division due to their thicker cell wall . However , the existence of the cyanobacterial genus Trichodesmium , in which cells perform nitrogen fixation and are capable of cell division [11] , [12] , suggests the possibility of other explanations . The general question of why selection has favored the evolution of multicellularity and cell differentiation has been explored in many previous studies [20]–[28] . The evolution of multicellularity is faced with a similar conflict as the evolution of cooperation in social organisms . The conflict arises because natural selection favors the propagation of individual's with the highest fitness , while the evolution and maintenance of cooperation requires selection to favor individuals with a behavior that incurs them a cost in fitness while increasing the fitness of other individuals . In this scenario , individuals with a non-cooperating phenotype or cheaters would reap the benefits paying none of the costs and be therefore the most fit . Indeed , cooperation has been shown to only arise when the fitness cost to an individual is outweighed by the benefits conferred on related individuals , a concept explained by inclusive fitness theory [29] . Many forms of conflict mediation have been proposed to facilitate the maintenance of cooperation in multicellular organisms [20] , [25] , [26] . Unicellular bottlenecks and small propagule size are believed to be a main factor in the maintenance of cooperation by ensuring that new organisms are composed of highly related cells . Once multicellularity evolves , cell differentiation and specialization can evolve , because it provides an increase in fitness to a group of related cells , which would not be otherwise possible [22] , [26] , [28] . Of all types of specialization , terminal differentiation , where cells lose their ability to reproduce new organisms may be the most extreme case of specialization . Whether cells are reversibly or terminally differentiated , they must always cooperate if the organism is to survive . Given that cells must cooperate , why a specific cell type evolves to become the germline while others evolve to become the soma is a topic that has received little attention . One proposed explanation is that differentiation can autonomously arise as a result of stochastic chemical interactions within and between cells [30] , and this can lead to terminal differentiation [31] . While this is a plausible hypothesis , it does not address the question of what evolutionary forces drive the evolution of different differentiation schemes such as reversible and terminal differentiation . Using a spatially explicit approach , we model here the evolution of differentiation . Our model follows assumptions about multicellular cyanobacterial species , but is nonetheless sufficiently general to apply to other systems . We assume that the physiological interactions of cells with neighbouring cells affects their reproductive success . We find that the topology of interactions , the differentiation costs , and the relative division rate between different cell types can all play a role in the evolution of terminal or reversible differentiation . In addition , we find that some conditions can lead to the “speciation” of a multicellular organism into a symbiotic pair of organisms . In this case , the different cell types separate into two lineages evolving independently from each other . Our approach helps to identify some of the principal factors that led to the evolution of the diverse differentiation strategies seen in simple multicellular organisms , such as the cyanobacteria .
In the model , every cell is characterised by four evolvable traits ( , , , ) which may have any value in the range ( Fig . 1A ) . Of these four traits , two traits ( , ) affect only photosynthetic cells , while the other two ( , ) affect only nitrogen fixing cells . The traits or control how much of the resources produced by a cell are kept for its own growth and division , while the remaining fraction or is given away to neighbouring cells . This means that a photosynthetic cell having a trait value of will keep all produced carbohydrates for its own cell growth while another cell with gives away all its produced carbohydrates . The traits or control the fraction of cells that differentiate into the other cell type immediately following a cell division . In other words , cells do not differentiate if they do not divide previously . For example , if a photosynthetic cell has the trait value , then 10% of its offspring cells will differentiate into nitrogen fixing cells . The individuals in our simulations evolve through mutation . This can occur every time a cell reproduces , at which time , traits in the daughter cell may mutate with probability , changing by a random amount that is uniformly distributed in the range . Cell composition ratios of carbon to nitrogen ( C∶N ) have been estimated to be around 6∶1 for bacterioplankton [32] . Typical sugar molecules produced in photosynthesis contain 6 carbon atoms . Therefore we consider the biomass composition to be 1 unit of carbohydrates to 1 unit of fixed nitrogen . Assuming that this ratio remains constant in the cell , and therefore that cells require carbohydrates and fixed nitrogen in equal parts , their division rate will be limited by the least available resource . Cell reproductive fitness is determined by division rate in the model . A cell's division rate depends on the amount of carbohydrates and fixed nitrogen available for its reproduction . Given these considerations , we define the fitnesses of a photosynthetic cell and of a nitrogen fixing cell as ( 1 ) ( 2 ) Here is a small constant that represents the base fitness and serves only to prevent the fitness from being zero , and is a parameter that determines how fast a photosynthetic cell divides relative to a nitrogen fixing cell given the same amount of resources . Differences in cell division rate between cell types can result from differences in cell composition , cell size [33] , the rate of biomass production , maintenance costs [34] , and regulatory effects . In the case of a nitrogen fixing cell , the total amount of carbohydrates or fixed nitrogen available to a cell for growth and division will be the fraction of received carbohydrates kept for reproduction minus the fraction consumed in nitrogen fixation to supply the cell with fixed nitrogen for its own growth . ( 3 ) ( 4 ) Here is the ratio that defines the amount of fixed nitrogen produced per carbohydrate consumed . The energetic costs of fixing nitrogen have been estimated to be 1 to 2 molecules of sugar for one molecule of ammonia [27] , [35] . For simplicity , we have assumed . To further simplify the model we assume that the nitrogen fixing cell is capable of regulating the amount of carbohydrates that it needs to consume for nitrogen fixation in order to achieve optimal growth . The optimal value of will depend on the ratio of carbohydrates to fixed nitrogen and should be given the 1∶1 ratio assumed here . This leads to the following fitness function for the nitrogen fixing cells: ( 5 ) In the case of a photosynthetic cell , the amount of carbohydrates available for its growth will be the fraction of produced carbohydrates kept for growth . The amount of fixed nitrogen available will be equal to the amount of fixed nitrogen received from neighboring nitrogen fixing cells: ( 6 ) ( 7 ) Here we assume that all photosynthetic cells produce one unit of carbohydrates . This leads to the following fitness function for the photosyntetic cells: ( 8 ) The amount of resources received from other cells will depend on many factors , such as the cell interaction topology , the interaction range , and the traits of the other cells ( Fig . 1 ) . is the amount of sugar received by nitrogen fixing cell , where ( 9 ) is the number of photosynthetic cells interacting with cell and is the number of nitrogen fixing cells interacting with cell . Meanwhile , is the amount of fixed nitrogen received by photosynthetic cell from interacting nitrogen fixing cells , where ( 10 ) The range of values and can have will depend on the interaction range , the type and trait values of neighboring cells . For example , when K = 4 , a nitrogen fixing cell could receive at most units of carbohydrates if it had 4 neighboring photosynthetic cells that gave away all their carbohydrates . To study the effects of differentiation costs we have modeled them as a reduction in the fitness of a differentiated cell by a fraction , such that the fitness of the cell becomes . After the first time a cell is chosen for division , this cost is removed . A differentiation cost modeled this way is equivalent to a reduction in resources available for growth by a proportional amount . Using a constant amount instead of a factor does not qualitatively change the results , as is shown in the results . Differentiation costs are expected to exist in differentiating cells because differentiation requires a cell to degrade the proteins expressed in its previous cell type . The degradation of these proteins therefore incurs a cost of energy or materials . It is also known to incur costs in higher organisms [36] . Fitness in our model is translated into a proportional probability that a cell will be chosen for reproduction every iteration . This probability of division is given by , where is the fitness of cell and is the sum over all fitness values of the cells in the population . In this model , cells are arranged in linear chains . When a cell reproduces , a new cell with the same traits is inserted in the chain between its parent and a neighbour ( Fig . 1B , C ) . We investigate two filament topologies that result as a consequence of the type of cell death considered . In the connected topology ( Fig . 1B ) , a cell chosen for death is simply removed from the chain , with one of the neighbours taking the place of the removed cell . In the broken chain topology ( Fig . 1C ) , the chain is broken in two parts when a cell chosen for death is removed , hence separating some of the neighbours of the removed cell . In addition , we study the effects of varying the distances at which fixed nitrogen and fixed carbon are exchanged by changing the interaction range between cells ( Fig . 1D ) . The interaction range represents the distance that nutrients are allowed to diffuse between cells due to proximity or through the transport of nutrients by vascular systems . Here we have considered the use of constant interaction strengths between cells , and vary instead only the number of neighboring cells that a cell can reach or interact with . We also investigate the use of an interaction strength defined by a Gaussian function which is presented in the supplementary information . The four traits ( , , , ) can evolve through mutation and selection to arrive at different sets of values . For the population to be viable both nitrogen fixing cells and photosynthetic cells must exist and exchange resources . This restriction implies that some sets of values such as ( , , ) can never evolve because no nitrogen fixing cells would be produced in a homogenous population of cells with this genotype . Fig . 2 provides a classification for the 6 genotypes that can evolve , which we will refer to as developmental strategies . Two developmental strategies correspond to terminally differentiating genotypes , where the nitrogen fixing cell is terminally differentiated ( I , violet ) , or where the photosynthetic cell is terminally differentiated ( VI , red ) . Two strategies are intermediate cases of terminal differentiation , where differentiated cells still divide ( II , blue; V , orange ) . One strategy corresponds to reversible differentiation where both cells can differentiate into the other cell type ( III , green ) . The last strategy corresponds to symbiosis , where both cells reproduce but do not differentiate ( IV , yellow ) . With this approach , a photosynthetic cell in the model can evolve from germline to soma , if the conditions imposed in the model favor that transition through mutation and selection . In this manner , we investigate the conditions that favor the evolution of the different developmental strategies .
We analysed the evolution of the variable traits ( , , , ) in populations of 400 cells starting with the set of initial trait values ( , , , ) . At the start of the simulation , all photosynthetic and nitrogen fixing cells are homogeneous with respect to their traits . Cells were initially placed in a single filament with periodic boundary conditions and randomly assigned as photosynthetic or nitrogen fixing with equal probability . The four panels in Fig . 3 show examples of the evolution of the population average of each trait in four different conditions . Each generation corresponds to 400 cell deaths and divisions . Instead of the differentiation rates and , the products and are plotted , because these express the effective rate of differentiation after cell division . For all simulations in Fig . 3 , it can be seen that the average trait values ( , , , ) evolve rapidly to a point were they begin fluctuating around a state which depends on the parameters of the simulation . The parameters investigated are the relative division rate , differentiation cost , filament topology , and interaction range . Using the averages of the variable traits we classify the evolved developmental strategy of the population according to Fig . 2 . For the purpose of classification , we consider trait values below the threshold of 0 . 05 to be effectively 0 . Fig . 3A shows the evolution of the averages of variable traits ( , , , ) over 5000 generation in an evolving population using a broken chain topology , where photosynthetic cells have a relative division rate three times faster ( ) than nitrogen fixing cells , and with no differentiation costs ( ) . We can see that in the final generation , photosynthetic cells keep half of the produced carbohydrates ( ) for their own cell growth and division , and differentiate at a rate of ( ) , while the nitrogen fixing cells do not keep any fixed nitrogen ( ) and therefore do not divide nor differentiate ( ) . Using Fig . 1 we classify this strategy as terminal differentiation with a photosynthetic germline ( I ) . Fig . 3B shows a simulation in the same conditions as in Fig . 3A , except that the photosynthetic cells divide three times more slowly ( ) . In this case we observe that the final strategy is terminal differentiation with a nitrogen fixing germline ( VI ) . Figs . 3C and 2D show simulations in the connected topology with slightly faster dividing photosynthetic cells ( ) . In Fig . 3C there are no differentiation costs ( ) and the final strategy corresponds to reversible differentiation ( III ) . In Fig . 3D there is a differentiation cost ( ) and the final strategy corresponds to the case of symbiosis ( IV ) ( the different cell types evolve into separate lineages ) . Next we investigate the sensitivity of the evolved developmental strategy to the initial traits ( , , , ) and whether different developmental strategies may evolve in the same conditions . In Fig . 4 , the solid lines show the plots of frequencies of the evolution of each developmental strategy when 50 stochastic simulations are carried out starting the simulation from a homogeneous population with initial traits ( , , , ) . In contrast , based on random initial conditions , the data points and error bars in Fig . 4 show the average and 95% confidence intervals for the frequency of evolved developmental strategies , respectively . This is estimated using bootstrapping from 500 simulations with random initial traits ( see supplementary information for details ) . Each plot shows how the frequencies change with varying relative division rate . The panels on the top ( Fig . 4A , B ) show the results in the case of the broken chain topology with no differentiation costs ( ) . Figs . 4C , D show the case of the connected topology with differentiation costs ( ) . Simulations for two different cell interaction ranges ( ) are shown in Fig . 4 . Other parameter combinations are shown in Fig . S1 and discussed in Text S1 . Different mutation rates ( ) and population sizes ( ) were tested and found to only change the number of generations needed for the system to evolve to the final developmental strategy . Lower mutation rates and larger population sizes required more generations for the population to reach the equilibrium compared to higher mutation rates or smaller population sizes . The confidence intervals observed in Fig . 4 are narrow , indicating that the developmental strategies which evolve are rather insensitive to the trait values of the initial population . Only a single strategy is generally seen to evolve under a set of conditions . However , at the points where a transition is observed between the most frequent strategies , two or more strategies evolve at appreciable frequencies , and that coincide with broader confidence intervals . For example , at in Fig . 4A ( broken chain topology , ) , a transition in the most frequently evolved strategy can be seen between terminal differentiation with nitrogen fixing germline ( VI , red ) and reversible differentiation ( III , green ) . At slightly larger than 1 in Fig . 4D ( connected topology , ) , many strategies can be seen to evolve with some frequency . At large differences in division rates ( or ) , when one cell divides much faster than the other , terminal differentiation without somatic division ( I , violet and VI , red ) evolves . Furthermore , it is the faster dividing cell type that becomes the germline . Hence , at low relative division rates ( ) , when nitrogen fixing cells are dividing faster , terminal differentiation with a nitrogen fixing germline ( VI , red ) is the most frequently evolved strategy . Conversely , at high relative division rates ( ) , when photosynthetic cells are the more rapidly dividing cells , terminal differentiation with a photosynthetic germline ( I , violet ) is the most frequently evolved strategy . To further examine the conditions which determine the most frequently evolved developmental strategies , we performed simulations for different relative division rates ranging from to , interaction ranges ranging from to , two different filament topologies ( broken chain and connected ) , and two values of differentiation costs ( and ) . Fig . 5 shows the most frequently evolved strategies ( represented as colours classified in Fig . 5A ) for each combination of parameters and simulated 50 times starting with initial traits ( , , , ) . All cases confirm that terminal differentiation ( I , violet and VI , red ) evolves at the extremes of relative division rate , in which the fastest dividing cell becomes the germline . In the broken chain topology , both with no differentiation costs ( Fig . 5A ) and with differentiation costs ( Fig . 5C ) , only three developmental strategies evolve . These are differentiation with a photosynthetic germline ( I , violet ) , reversible differentiation ( III , green ) , and terminal differentiation with a nitrogen fixing germline ( VI , red ) . In both cases it can be seen that the main factor influencing the evolved developmental strategy is the relative division rate ( ) , with little dependency on the interaction range of the cells ( ) . In Fig . 5A , where no differentiation costs were included , fast dividing photosynthetic cells ( ) result in the evolution of terminal differentiation with photosynthetic cells as the germline ( I , violet ) . Slow dividing photosynthetic cells ( ) also lead to the evolution of terminal differentiation , but in this case the nitrogen fixing cells become the germline ( VI , red ) . For intermediate relative division rates ( ) , reversible differentiation ( III , green ) is the evolved strategy . When a differentiation cost is considered ( Fig . 5C ) , the range under which reversible differentiation ( III , green ) evolves is limited to at low values . Conversely , the range of values under which terminal differentiation ( I , violet and VI , red ) evolves increases . For the connected topology with no differentiation costs ( Fig . 5B ) , the result is qualitatively similar to the one observed for the broken chain topology with ( Fig . 5A ) . In both cases only three strategies are observed to evolve most frequently , the two types of terminal differentiation without somatic division ( I , violet and VI , red ) and reversible differentiation ( III , green ) . Remarkably , with differentiation costs and a connected topology ( Fig . 5D ) , all developmental strategies evolve in some range of conditions . Reversible differentiation ( III , green ) is reduced to a very narrow range of conditions with intermediate values of interaction ranges ( ) and slightly faster dividing photosynthetic cells ( ) . When , the range of conditions previously occupied by reversible differentiation ( III , green ) is replaced by terminal differentiation with somatic division ( II , blue and V , orange ) at shorter interaction ranges ( ) , and symbiosis ( IV , yellow ) at longer interaction ranges ( ) . One assumption we have made that may not apply to other systems is that nitrogen fixing cells are only able to fix nitrogen if they obtain carbohydrates from photosynthetic cells . This results in an asymmetry in the model because photosynthetic cells do not require fixed nitrogen to perform photosynthesis , though they require it for cell growth and division . We show in Fig . S4 that the results presented here do not qualitatively change when we modify the model to enable nitrogen fixing cells to fix nitrogen independently of the carbohydrates received . We also explored other modifications and found that in all cases the results have remained qualitatively the same . In Fig . S5 we show the results of using a constant differentiation cost instead of a cost that decreases the resources available to a cell by a fraction . In Fig . S6 , we show the results when using a Gaussian function to define the interaction strengths between cells .
The results shown here establish a strong link between the relative division rate of different cell types and the cell type that becomes the germline in a multicellular organism . Figs . 4 and 5 indicate that when one cell type divides faster than the other , it evolves to become the germline . This result is found to be independent of the differentiation costs ( ) , filament topology , and interaction range ( ) . The reason can be explained intuitively by noting that an organism that requires both cell types will divide faster when the fastest dividing cell type is the one that produces the other cell type as needed . Hence the faster dividing cell types are the ones which remain pluripotent . For example this pattern is seen in plants , where cells in the apical meristems generating shoots and roots consist of rapidly dividing undifferentiated cells [37] , [38] . Equivalently , one can interpret this as a situation in which cells that have a higher fitness at the individual level are the ones that become the germline . When division rates of the different cell types are comparable and , our model shows that reversible differentiation ( III , green ) evolves ( Figs . 5A and 5B ) . This corresponds to the case of differentiated cells that have the ability to de-differentiate into another cell type . Examples exist in many plants and in some animals capable of regeneration [5] , [6] . Although terminal differentiation is found to evolve in the widest range of conditions , reversible differentiation can evolve in conditions where the division rates of different cell types are comparable . The latter can happen even in the absence of selection for the ability to regenerate or reproduce by fragmentation ( Figs . 5B , D ) . It is important to note that large differences in cellular division rates are a necessary but insufficient condition for a cell type to become the germline . The fast growth rate of a cell type must not harm the fitness of the organism as a whole , otherwise faster growing cells such as cancer cells would become the germline more often . Such an eventuality has occurred only on rare occasions [39] , [40] . Cell interaction affects developmental strategies in two ways . First , the broken chain topology ( Fig . 5A ) increases the range of conditions under which reversible differentiation ( III , green ) evolves when compared to the connected topology ( Fig . 5B ) . The reason can be understood if we consider that reversible differentiation increases the survival of filaments in response to fragmentation . By ensuring that either cell type can produce the other cell type , the probability that a fragment will carry only non-differentiating cells is reduced . A similar argument can be made to explain why symbiosis ( IV , yellow ) does not evolve in the broken chain topologies under any conditions ( Figs . 5A and 5C ) . In these topologies , broken fragments never come into contact again , meaning that once a symbiotic pair within a filament is split , it will be condemned to death . Hence , such mutants can never become fixed in the population . The effect of interaction range ( ) is mainly seen in connected topologies . In this case , all possible developmental strategies evolve in at least one set of conditions ( Fig . 5B , D ) . For example , the symbiotic state ( IV , yellow ) that was not found in broken chain topologies , occurs in the connected topology if interaction ranges ( ) are sufficiently high ( ) and if there are differentiation costs ( ) . In the case with differentiation costs , increasing the interaction range leads to a decrease in the range of relative division rates under which terminal differentiation evolves , while the range for other strategies expands ( Fig . 5D ) . Higher interaction ranges ( ) in the connected topology are shown in Fig . S3 and discussed in Text S1 . They lead to a slight increase in the range of relative division rates in which symbiosis ( IV , yellow ) and terminal differentiation with a nitrogen fixing germline and somatic division ( V , orange ) occur . It is well known that topologies with few interactions promote cooperative behaviour , while fully connected topologies , where all individuals interact with each other , result in the invasion of cheaters [41] , [42] . This has already been shown to be the case in a model of cyanobacteria [27] , in which populations of vegetative and heterocyst cells are driven to extinction in the fully connected case . Here , we have analysed topologies that are far from the fully connected case , and where several forms of cooperation are stable . By varying the relative division rate , several developmental strategies such as reversible differentiation and symbiosis can evolve in the same filament topology and interaction range ( Fig . 5 ) . These developmental strategies are neither altruistic nor selfish , since both cell types can divide . Hence , the mapping of our present results to established concepts in social biology may require further work . Multicellular cyanobacteria have evolved several of the developmental strategies seen in this model . Terminally differentiating cyanobacteria such as Anabaena or Nostoc have filamentous forms composed of two different cell types: vegetative cells that are photosynthetic , divide and differentiate into the other cell type , and heterocyst cells that fix nitrogen and are unable to divide . The latter can be distinguished by their larger size and thicker cell walls [8] . Our model provides clues to why heterocystous cyanobacteria form terminally differentiated heterocysts that do not divide . An ad-hoc explanation based on a proximal cause is that a heterocyst's thicker cell wall impedes it from undergoing cell division . However , our results provide an alternative explanation . In light of the model , a thicker cell wall corresponds to added costs and therefore a slower division rate . Under this condition , the developmental strategy that maximises the organism's fitness is terminal differentiation without somatic division ( I , violet and VI , red ) ( Fig . 5C ) . This means that the ultimate reason why heterocysts do not divide is not necessarily due to mechanistic constraints , but rather a result of evolutionary constraints . The only known example of potentially reversibly differentiated cyanobacteria is Trichodesmium . In species of this genus , different cell types are morphologically indistinguishable . However , differences at the level of expression of nitrogenase exist , and nitrogen fixation is shown to occur in distinct cells found across the filaments [11] . Although cells are differentiated in their expressed protein and function , both cell types maintain their ability to divide [43] , [44] . While no direct experiment has shown that cells in Trichodesmium reversibly differentiate , the fact that the fraction of nitrogen fixing cells varies with daily rhythmicity , reaching a maximum of 24% during the day and a minimum of 5% before dawn , suggests that the nitrogen fixing cells reversibly differentiate into photosynthetic cells [45] . In this case again , our results provide some insights as to why cells that are specialised in nitrogen fixation ( therefore similar to heterocysts ) are not terminally differentiated , but are still capable of dividing and of reverting back to a photosynthetic phenotype . Since both cell types are structurally similar , they can be expected to have similar division rates . The results shown in Figs . 4A and 5A predict that reversible differentiation ( III , green ) should be the most frequently evolved developmental strategy in this case . So far , no known examples of multicellular cyanobacteria exist in which terminally differentiating nitrogen fixing cells ( heterocysts ) are capable of cell division ( II , blue ) . While this can simply reflect our incomplete knowledge , our results suggest that such developmental strategies are evolutionarily unstable ( Fig . 5A–D ) . The finding that symbiosis evolves in a connected topology under several conditions of relative cell division rate and differentiation costs points to some interesting evolutionary possibilities . One is that some organisms may have speciated as a result of changing conditions that initially selected for terminal or reversible differentiation , but later changed to favour a symbiotic state . Potential support for this idea comes from a recently sequenced cyanobacterium named UCYN-A that is closely related to a member species of the genus Cyanothece [46] . Cyanothece are unicellular circadian cyanobacteria capable of photosynthesis and nitrogen fixation by temporally separating the two processes . The newly sequenced relative of Cyanothece lacks the genes necessary to perform photosynthesis found in Cyanothece species [46] . Instead , it has only the genes necessary for nitrogen fixation . Because it is unable to perform photosynthesis , it is dependent on obtaining its carbohydrates from the environment or from other organisms . This suggests that a scenario in which cyanobacteria speciate into symbiotic or interacting collectives is possible . In effect , chloroplasts , which are endosymbionts that descended from cyanobacteria , are a likely endpoint of such a scenario . In this case , chloroplasts provide the host plant with fixed carbon while the plant is the intermediary that provides fixed nitrogen . Plants have never evolved the ability to fix nitrogen . They absorb it from the environment or rely instead on symbiotic diazotrophic bacteria such as the cyanobacterium Nostoc to fix nitrogen in exchange for carbohydrates produced by the photosynthetic plant [13] . The vascular system of plants conceptually changes the topology of cell interactions from a chain to a connected topology with high interaction ranges , allowing photosynthetic plant cells to exchange nutrients with the nitrogen fixing cyanobacteria in the roots of the plant . Our results show that in such conditions ( Figs . 4D and 5D ) , a symbiotic relationship ( IV , yellow ) where the nitrogen fixing cells evolve independently from the photosynthetic cells is a frequently evolved strategy . The range of values in which symbiosis evolves is seen to increase with higher differentiation costs ( Fig . S2 and Text S1 ) and interaction ranges ( Fig . S3 and Text S1 ) . These results suggest that the symbiotic relationship between plants and cyanobacteria may be evolutionarily more stable than the alternative scenario , in which plants would fix their own nitrogen . While this model draws inspiration from differentiated cyanobacteria , the results found here may apply to a wider range of biological systems . In essence , the model describes the evolution of a simple multicellular organism or population with two types of individuals that produce different resources , but require both resources to reproduce . Hence , these individuals need to interact and exchange resources . By considering the exchange of fitness benefits as a form of resource exchange , a cell type in an organism that serves a structural function can also be analysed using such a model . In the supplementary information ( Text S1 ) we present the results of several modifications to the model which do not qualitatively change the results found . The modifications we considered comprise a nitrogen fixing cells that do not need carbohydrates to fix nitrogen ( Fig . S4 ) , a fixed differentiation cost instead of a fractional cost ( Fig . S5 ) , and a Gaussian function to describe the interaction strengths ( Fig . S6 ) . In all cases we found that faster dividing cells evolve to become the germline , and all developmental strategies can evolve in some range of conditions . These results lend support to the idea that the observations we made do not just apply to cyanobacteria but can apply to a range of other simple differentiated multicellular organisms . This model shows that in simple organisms , the optimum developmental strategy depends on how cells divide and interact . We have shown that the topology of interactions , the interaction range , the differentiation costs , and the relative division rate between cell types all play a role in the type of differentiation that evolves . However , the difference in cell division rates is the main factor determining the type of differentiation that evolves . Furthermore , it determines the cell type that becomes the germline . Hence , we establish for the first time the conditions that drive the evolution of terminal and reversible differentiation . | The evolution of multicellularity is one of the most fascinating topics of evolutionary biology . Without multicellularity the incredible diversity of extant life would not be possible . In many multicellular organisms with specialized cells , some cell types become terminally differentiated ( somatic cells ) and lose the ability to reproduce new organisms while other cells maintain this ability ( germline ) . Little is known about the conditions that favor the evolution of terminal differentiation in multicellular organisms . To understand this problem we have developed a computational model , inspired by multicellular cyanobacteria , in which the cells in an organism composed of two cell types ( photosynthetic and nitrogen fixing ) are allowed to evolve from germline to soma cells . We find three striking results . First , faster dividing cell types always evolve to become the germline . Second , the conditions under which we find different outcomes from the model are in good agreement with the different forms of development observed in multicellular cyanobacteria . Third , some conditions lead to a symbiotic state in which the two cell types separate into different lineages evolving independently of one another . Remarkably , cyanobacteria are also known to engage in symbiotic relationships with plants , producing fixed nitrogen for the plant in exchange for carbohydrates . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"organismal",
"evolution",
"theoretical",
"biology",
"microbial",
"evolution",
"biology",
"evolutionary",
"biology",
"microbiology",
"evolutionary",
"theory",
"evolutionary",
"developmental",
"biology"
] | 2012 | Differences in Cell Division Rates Drive the Evolution of Terminal Differentiation in Microbes |
Pre–mRNAs are often processed in complex patterns in tissue-specific manners to produce a variety of protein isoforms from single genes . However , mechanisms orchestrating the processing of the entire transcript are not well understood . Muscle-specific alternative pre–mRNA processing of the unc-60 gene in Caenorhabditis elegans , encoding two tissue-specific isoforms of ADF/cofilin with distinct biochemical properties in regulating actin organization , provides an excellent in vivo model of complex and tissue-specific pre–mRNA processing; it consists of a single first exon and two separate series of downstream exons . Here we visualize the complex muscle-specific processing pattern of the unc-60 pre–mRNA with asymmetric fluorescence reporter minigenes . By disrupting juxtaposed CUAAC repeats and UGUGUG stretch in intron 1A , we demonstrate that these elements are required for retaining intron 1A , as well as for switching the processing patterns of the entire pre–mRNA from non-muscle-type to muscle-type . Mutations in genes encoding muscle-specific RNA–binding proteins ASD-2 and SUP-12 turned the colour of the unc-60 reporter worms . ASD-2 and SUP-12 proteins specifically and cooperatively bind to CUAAC repeats and UGUGUG stretch in intron 1A , respectively , to form a ternary complex in vitro . Immunohistochemical staining and RT–PCR analyses demonstrate that ASD-2 and SUP-12 are also required for switching the processing patterns of the endogenous unc-60 pre-mRNA from UNC-60A to UNC-60B in muscles . Furthermore , systematic analyses of partially spliced RNAs reveal the actual orders of intron removal for distinct mRNA isoforms . Taken together , our results demonstrate that muscle-specific splicing factors ASD-2 and SUP-12 cooperatively promote muscle-specific processing of the unc-60 gene , and provide insight into the mechanisms of complex pre-mRNA processing; combinatorial regulation of a single splice site by two tissue-specific splicing regulators determines the binary fate of the entire transcript .
Alternative pre-mRNA processing is a major way to produce a number of different mRNAs and proteins from one gene [1] , [2] . Recent transcriptome analyses by deep sequencing estimated that more than 90% of human multi-exon genes undergo alternative processing and most alternative processing events are regulated in tissue-specific manners [3] , [4] . These alternative pre-mRNA processing events are classified into seven elementary events: cassette exons , mutually exclusive exons , alternative 5′ splice sites , alternative 3′ splice sites , intron retention , alternative first exons and alternative polyadenylation sites [5] , [6] . A variety of tissue-specific splicing factors and RNA secondary structures have been shown to regulate these elementary events in the minigene context or by knockdown and/or knockout experiments [7] , [8] , [9] . However , pre-mRNA processing in multicellular organisms is often complex due to various combinations of the elementary events and the molecular mechanisms by which tissue-specific factors regulate such complex alternative processing of the entire gene in vivo remain to be elucidated . Muscle is one of tissues in which many genes undergo tissue-specific pre-mRNA processing [3] , [4] . A number of muscle-specific protein isoforms are expressed by alternative pre-mRNA splicing and play adapted roles depending on the specific properties of muscle fiber types [10] , [11] , [12] . For instance , tissue-specific splicing generates functionally distinct isoforms of tropomyosin [13] and troponin T [14] . Global analyses of splicing patterns during development of heart and skeletal muscle revealed that splicing transitions of these genes occur at specific times [15] , [16] . Bioinformatics analyses have revealed candidate cis-elements regulating muscle-specific splicing patterns [16] , [17] , [18] . In addition , several trans-acting splicing factors are known to regulate muscle-specific alternative splicing . These include muscleblind-like ( MBNL ) [19] , RBFOX family [20] , CUGBP and ETR-3 like factor ( CELF ) family [21] , polypyrimidine tract binding protein ( PTB ) [22] and hnRNP H [23] . However , how multiple splicing factors coordinate regulation of specific splicing events is poorly understood . Alternative processing of the uncoordinated ( unc ) -60 gene in Caenorhabditis elegans provides an excellent model of muscle-specific and complex pre-mRNA processing of genes related to contractile apparatuses . The unc-60 gene encodes two homologous proteins , UNC-60A and UNC-60B [24] , which are members of the actin depolymerising factor ( ADF ) /cofilin family of actin-binding proteins that promote rapid turnover of the actin cytoskeleton [25] . The unc-60 gene consists of a common first exon and two separate series of downstream exons , exons 2A through 5A for UNC-60A and exons 2B through 5B for UNC-60B ( Figure 1A ) . Alternative choices of exons 2A–5A or exons 2B–5B result in tissue-specific expression patterns of the two ADF/cofilin isoforms: UNC-60A protein is expressed in most embryonic cells throughout embryogenesis and predominantly expressed in non-muscle tissues , while UNC-60B protein is mainly detected in body wall muscles [26] . Our biochemical and genetic studies demonstrated that the UNC-60 isoforms have distinct biochemical properties in the regulation of actin dynamics [27] , [28] and different in vivo functions during development and in muscle organization [26] , [29] . The structure of the unc-60 gene and its expression patterns raise a question as to how the first exon and the two series of downstream exons are properly spliced in a tissue-specific manner . We previously reported genetic evidence that an RNA-binding protein SUP-12 , which has only one RNA-recognition motif ( RRM ) , is required for generation of muscle-specific UNC-60B mRNA [30] . However , the molecular mechanism by which SUP-12 regulates the muscle-specific alternative processing of the unc-60 gene remains unclear . In this study , we applied a transgenic alternative splicing reporter system [31] , [32] , [33] to visualize muscle-specific alternative processing patterns of the unc-60 pre-mRNA . We demonstrate that repression of excision of the intron between exon 1 and exon 2A is the fate-determining event for the unc-60 transcript . We provide genetic and biochemical evidence that SUP-12 and another muscle-specific splicing regulator Alternative-Splicing-Defective-2 ( ASD-2 ) , a member of the signal transduction and activation of RNA ( STAR ) family of RNA-binding proteins [34] , cooperatively repress excision of the first intron through specific binding to the intron . Our data provide in vivo evidence that combinatorial regulation of a single splice site by two tissue-specific splicing regulators determine the binary fate of the entire transcript that can potentially be processed into two alternative isoforms .
In order to visualize the binary processing patterns of the unc-60 transcript in vivo , we intended to construct a pair of fluorescence alternative processing reporter minigenes . If the intron between exon 1 and exon 2A ( hereafter called intron 1A ) is excised prior to selection of exon 2B , it would be impossible to produce UNC-60B mRNA . We therefore assumed that excision of intron 1A should be repressed until exon 2B is transcribed in tissues where UNC-60B is expressed . On the basis of the assumption , we constructed an asymmetric pair of reporter minigenes , unc-60E1-E2A-RFP and unc-60E1-E3B-GFP . The unc-60E1-E2A-RFP cassette , carrying unc-60 genomic DNA fragment from exon 1 through exon 2A ( Figure 1B , top panel ) , was designed to monitor excision of intron 1A via expression of RFP-fusion protein ( UNC-60A-RFP ) . If intron 1A is retained ( UNC-60-I1A ) , RFP would not be expressed due to a premature termination codon in intron 1A ( Figure 1B , top panel ) . On the other hand , the unc-60E1-E3B-GFP cassette , carrying unc-60 genomic DNA fragment from exon 1 through exon 3B ( Figure 1B , bottom panel ) , was designed to monitor UNC-60B-type processing via expression of GFP-fusion protein ( UNC-60B-GFP ) . An intact UNC-60A isoform ( UNC-60A-full ) would be expressed in tissues where UNC-60A is expressed ( Figure 1B , bottom panel ) . We successfully visualized the alternative expression of the UNC-60 isoforms with the unc-60 reporter cassettes under the control of the unc-51 promoter that directs expression in a broad variety of tissues [35] , [36] . As expected , the expression patterns of UNC-60A-RFP and UNC-60B-GFP varied between muscle and non-muscle tissues ( Figure 1C , 1D ) . Non-muscle tissues including the nervous system and intestine expressed UNC-60A-RFP ( Figure 1C , 1D , left panels ) , and muscle tissues such as body wall muscles and pharyngeal muscles expressed UNC-60B-GFP ( Figure 1C , 1D , right panels ) . This result is consistent with our previous immunohistochemical studies showing that UNC-60A and UNC-60B proteins were detected in non-muscle and muscle tissues , respectively [26] , [37] . We checked splicing patterns of mRNAs derived from the unc-60 reporter cassettes by cloning and sequencing reverse transcription-polymerase chain reaction ( RT-PCR ) products , and confirmed that the four mRNA isoforms schematically shown in Figure 1B were actually generated in the transgenic worms ( data not shown ) . To focus on the muscle-specific control of the unc-60 processing , we utilized myo-3 promoter to drive expression of the unc-60 reporter specifically in body wall muscles . Transgenic worms with an integrated transgene allele ybIs1831 [myo-3::unc-60E1-E2A-RFP myo-3::unc-60E1-E3B-GFP] predominantly expressed UNC-60B-GFP in body wall muscles ( Figure 1E ) , consistent with the unc-60 reporter expression in muscles ( Figure 1C , 1D ) . We therefore used the myo-3 promoter for further analyses described below . To test whether muscle-specific repression of UNC-60A-RFP and expression of UNC-60B-GFP from the unc-60 reporter are similarly regulated by a muscle-specific splicing regulator SUP-12 to the endogenous mRNAs for UNC-60A and UNC-60B isoforms [30] , we crossed the reporter allele ybIs1831 with a presumptive null allele sup-12 ( yb1253 ) [38] . As expected , the reporter worms clearly turned the colour from Green to Red in the sup-12 background ( Figure 2A ) , confirming that SUP-12 is required for the muscle-specific expression profile of the unc-60 reporter . In a previous study , we identified SUP-12 as a co-regulator of mutually exclusive exons of a fibroblast growth factor receptor gene egg-laying-defective ( egl ) -15 [38] . In the case of repression of egl-15 exon 5B , SUP-12 functions as a muscle-specific partner of the Fox-1 family proteins ASD-1 and FOX-1 [31] , [38] . We therefore speculated that other regulator ( s ) may also be involved in the muscle-specific regulation of unc-60 . As direct interaction between SUP-12 and ASD-1 in a yeast two-hybrid system had been reported in a worm interactome study [39] , we screened for a putative co-regulator of the unc-60 reporter by knocking down genes encoding possible SUP-12-interactors ASD-1 , ASD-2 , ETR-1 , MEC-8 , R02F2 . 5 and W02A11 . 3 , deposited in the database ( http://interactome . dfci . harvard . edu/ ) . We performed RNA interference ( RNAi ) by feeding the reporter worms with bacterial clones targeting the six genes , and found that knockdown of asd-2 led to expression of UNC-60A-RFP ( Figure S1 ) . We previously identified ASD-2 , an RNA-binding protein belonging to the STAR family , as a regulator of muscle-specific and developmentally regulated alternative splicing of a collagen gene let-2 [32] , [33] . The asd-2 gene has alternative first exons and a non-lethal allele asd-2 ( yb1540 ) has a nonsense mutation in the asd-2b-specific first exon ( Figure 2B ) , which is used in body wall muscles and pharyngeal muscles [32] . The unc-60 reporter worms exhibited weak Red phenotype in the asd-2 ( yb1540 ) background ( Figure 2C ) and body wall muscle-specific expression of ASD-2b cDNA rescued the colour phenotype ( Figure 2D ) , confirming that asd-2b is involved in the muscle-specific regulation of the unc-60 reporter . To investigate subcellular localization of ASD-2 , we raised polyclonal antibodies against recombinant full-length ASD-2b protein and stained wild-type and asd-2 ( yb1540 ) worms with a purified immunoglobulin G ( IgG ) fraction ( Figure 2E , 2F ) . Nuclei of body wall muscles , which are aligned along the dorsal and ventral periphery , are stained in the wild type ( Figure 2E ) and not in asd-2 mutant ( Figure 2F ) . In Western blotting , the same antibody detected a major band with an apparent molecular weight of 56 kDa in wild-type and not in asd-2 ( yb1540 ) lysate ( Figure 2G ) . These results indicated that ASD-2b is the major isoform and is predominantly localized in the nuclei of body wall muscles . RNAi by micro-injecting double-stranded RNA ( dsRNA ) , a more effective method than feeding dsRNA-expressing bacteria , led to a stronger Red phenotype ( Figure 2C ) , suggesting trace remaining activity of ASD-2 in asd-2 ( yb1540 ) mutant . To confirm splicing patterns of mRNAs derived from the unc-60 reporter minigenes in body wall muscles , we performed RT-PCR analysis with minigene-specific primer sets ( Figure 2H ) . In the wild-type background , UNC-60B-type mRNA , UNC-60B-GFP , was predominantly generated from unc-60E1-E3B-GFP ( Figure 2H , middle panel , lane 1 ) . A transcript derived from unc-60E1-E2A-RFP was almost undetectable ( Figure 2H , top panel , lane 1 ) , presumably due to rapid degradation of a non-productive mRNA isoform , UNC-60-I1A , by nonsense-mediated mRNA decay ( NMD ) [40] . On the other hand , the amount of UNC-60B-GFP was reduced and UNC-60A-type mRNAs , UNC60A-RFP and UNC-60A-full , were detected in asd-2 and sup-12 mutants ( Figure 2H , lanes 2 and 3 ) , consistent with their colour phenotypes shown in Figure 2C and 2A , respectively . These results confirmed that both SUP-12 and ASD-2 are responsible for switching the processing patterns of the unc-60 reporter from UNC-60A-type to UNC-60B-type in body wall muscles . The experiments described above indicate that each of the unc-60 reporter minigenes , even the shorter one , carries sufficient regulatory elements for ASD-2 and SUP-12 to switch from non-muscle-type to muscle-type processing . As regulatory elements for alternative splicing are often evolutionarily conserved in introns among nematodes [31] , [32] , [38] , [41] , we searched for conserved stretches in unc-60 intron 1A in the Caenorhabditis genus . Alignment of nucleotide sequences available in WormBase ( http://www . wormbase . org/ ) revealed that CTAAC repeats and TGTGTG stretch are highly conserved just upstream of the splice acceptor site ( Figure 3A ) . To evaluate the roles of these elements in the muscle-specific processing of the unc-60 reporter , we constructed two pairs of modified unc-60 reporter minigenes M1 and M2 . In the M1 pair , CTAAC repeats were mutagenized to CAAAC ( Figure 3B ) . In the M2 pair , TGTGTG were mutagenized to TATATA ( Figure 3B ) . Disruption of either of the two elements resulted in Red phenotype ( Figure 3C ) , phenocopying sup-12 mutant ( Figure 2A ) and asd-2 ( RNAi ) worms ( Figure 2C ) . RT-PCR analysis of mRNAs derived from the mutant reporters revealed that both M1 and M2 mutations increased production of UNC-60A-RFP ( Figure 3D , top panel ) and decreased expression of UNC-60B-GFP ( Figure 3D , bottom panel ) , consistent with their colour phenotypes . These results confirmed that the colour phenotypes observed with the mutant reporters are due to altered patterns of pre-mRNA processing . We concluded that both CUAAC repeats and UGUGUG stretch are required for muscle-specific repression of intron 1A excision . Notably , expression of UNC-60A-full mRNA from the M1 and M2 mutants of unc-60E1-E3B-GFP minigene increased compared to the wild-type minigene ( Figure 3D ) , indicating that the repression of intron 1A excision via CUAAC repeats and UGUGUG stretch is a crucial event to switch the processing patterns of the entire unc-60E1-E3B-GFP minigene from UNC-60A-type to UNC-60B-type . To confirm direct and specific binding of ASD-2 and SUP-12 to the cis-elements in unc-60 intron 1A in vitro , we prepared radiolabelled RNA probes containing the intact sequence ( WT ) or those with mutations as in the mutant reporters ( M1 and M2 ) ( Figure 4A ) and recombinant full-length ASD-2b and full-length SUP-12 proteins ( Figure 4B ) to perform electrophoretic mobility shift assays ( EMSAs ) ( Figure 4C , 4D ) . Recombinant ASD-2b protein shifted the mobility of WT ( Figure 4C , lanes 1–6 ) and M2 ( Figure 4D , lanes 18–22 ) probes in a dose-dependent manner and not of M1 probe ( Figure 4D , lanes 1–5 ) , demonstrating direct and specific binding of ASD-2b to CUAAC repeats . On the other hand , recombinant SUP-12 protein shifted the mobility of WT ( Figure 4C , lanes 13–18 ) and M1 ( Figure 4D , lanes 6–9 ) probes to a similar extent in a dose-dependent manner and less efficiently of M2 probe ( Figure 4D , lanes 23–26 ) to a less extent , demonstrating direct and specific binding of SUP-12 to UGUGUG stretch . The result also indicated that SUP-12 could bind to other site ( s ) in the probes with a lower affinity . We next asked whether ASD-2b and SUP-12 cooperatively bind to unc-60 intron 1A RNA . We analyzed supershifts of the mobility of the unc-60 intron 1A probes by the combination of ASD-2b and SUP-12 in EMSAs ( Figure 4C , 4D ) . ASD-2b efficiently supershifted the mobility of WT probe at lower concentrations in the presence of SUP-12 ( Figure 4C , lanes 7–12 ) compared to ASD-2b alone ( lanes 1–6 ) . In the same way , SUP-12 supershifted the mobility of WT probe at lower concentrations in the presence of ASD-2b ( lanes 19–24 ) compared to SUP-12 alone ( lanes 13–18 ) . These results indicated that ASD-2b and SUP-12 cooperatively form a stable ASD-2b/SUP-12/RNA ternary complex with unc-60 intron 1A RNA . ASD-2b failed to supershift the mobility of M1 probe ( Figure 4D , lanes 10–17 ) , indicating that CUAAC repeats are essential for the ternary complex formation . SUP-12 less efficiently supershifted the mobility of M2 probe ( Figure 4D , lanes 31–34 ) compared to WT probe ( Figure 4C , lanes 21–24 ) in the presence of ASD-2b , indicating that UGUGUG stretch is involved in the ternary complex formation . We finally asked whether ASD-2b and SUP-12 can preform a complex in the absence of unc-60 intron 1A by pull-down experiments ( Figure 4E ) . Glutathione-S-transferase ( GST ) -fused full-length ASD-2b protein pulled down a substantial amount of recombinant full-length SUP-12 protein in the absence of target RNAs ( Figure 4E , lane 2 ) and wild-type ( WT ) unc-60 intron 1A ( unc-60-I1A ) RNA enhanced the pull-down efficiency in a dose-dependent manner ( lanes 3 , 4 ) . On the other hand , GST-fused monomeric RFP ( mRFP ) protein failed to pull down SUP-12 protein even in the presence of unc-60-I1A RNA ( lanes 10–13 ) , demonstrating specific interaction between ASD-2b and SUP-12 . M1 and M2 mutant unc-60-I1A RNAs less effectively enhanced the interaction between ASD-2b and SUP-12 ( lanes 5–8 ) , consistent with their weaker or no ability to form a ternary complex ( Figure 4D ) . We therefore concluded that ASD-2b and SUP-12 can weakly interact with each other and that unc-60 intron 1A RNA promotes the formation of the stable ASD-2b/SUP-12/RNA ternary complex by providing juxtaposed CUAAC repeats and UGUGUG stretch that are specifically recognized by ASD-2b and SUP-12 , respectively . We examined whether ASD-2 regulates muscle-specific pre-mRNA processing of the endogenous unc-60 gene . We have demonstrated that ASD-2 and SUP-12 cooperatively switch alternative processing of the unc-60 reporter from UNC-60A-type to UNC-60B-type in body wall muscles . If this model can be applied to the endogenous unc-60 gene , worms depleted of asd-2 function should ectopically express UNC-60A in place of UNC-60B in body wall muscles . Indeed , RT-PCR analysis of the endogenous UNC-60 mRNAs revealed that relative amount of UNC-60B mRNA was decreased in asd-2 ( yb1540 ) ; asd-2 ( RNAi ) worms ( Figure S2 ) . To further test the splicing change in body wall muscles , we investigated expression of UNC-60A protein by immunohistochemistry ( Figure 5A , 5B ) . In wild-type worms , UNC-60A was undetectable in body wall muscles ( Figure 5A , encircled ) but was detected in other tissues ( Figure 5A , left panel ) . Knockdown of the asd-2 gene resulted in ectopic expression of UNC-60A in body wall muscles ( Figure 5B , encircled ) , confirming that ASD-2 determines the processing patterns of the endogenous unc-60 gene in body wall muscles . Our previous work demonstrated that sup-12 mutation strongly suppressed structural defects of body wall muscles and paralysis of UNC-60B-specific mutant , unc-60B ( su158 ) [30] . The deletion allele su158 lacks exons 3B and 4B ( Figure 1A ) , and suppression of the phenotypes by sup-12 mutation was likely due to ectopic expression of UNC-60A [30] . We therefore investigated whether knockdown of the asd-2 gene also suppresses phenotypes of unc-60B ( su158 ) mutant . Wild-type worms exhibited sinusoidal locomotion ( Figure 5C , left panel ) , and actin filaments were organized in a striated pattern ( Figure 5C , right panel ) . On the other hand , unc-60B ( su158 ) worms were almost paralyzed ( Figure 5D , left panel ) with severe disorganization of actin filaments ( Figure 5D , right panel ) . We found that asd-2 ( yb1540 ) ; unc-60B ( su158 ) double mutant slightly restored motility and actin filament organization ( Figure 5E ) . Since asd-2 ( RNAi ) worms showed a severer colour phenotype than asd-2 ( yb1540 ) allele ( Figure 2C ) , we further knocked down remaining activity of asd-2 by RNAi . As expected , asd-2 ( yb1540 ) ; unc-60B ( su158 ) ; asd-2 ( RNAi ) worms restored sinusoidal locomotion ( Figure 5F , left panel ) and actin filament organization was greatly improved ( Figure 5F , right panel ) . We confirmed by immunohistochemistry that asd-2 ( yb1540 ) mutation and/or asd-2 ( RNAi ) resulted in ectopic expression of UNC-60A in body wall muscles in the unc-60B ( su158 ) background ( Figure S3 ) . Transgenic expression of UNC-60A ( Figure 5G ) as well as UNC-60B ( Figure 5H ) in body wall muscles restored sinusoidal locomotion of unc-60B ( su158 ) mutant , indicating that UNC-60A can exert , at least in part , functions of muscle-specific UNC-60B isoform and that possible splicing change in other genes are not required for the phenotype suppression . These observations demonstrate that ASD-2 is a bona fide regulator of the muscle-specific pre-mRNA processing of the endogenous unc-60 gene as well as SUP-12 . Finally , we analyzed splicing patterns of mature and partially spliced RNAs from the endogenous unc-60 gene ( Figure 6 ) . For this experiment , we used wild-type and sup-12 ( yb1253 ) worms because asd-2 ( yb1540 ) mutation exhibited weaker effect on the unc-60 reporter . In the wild type , mature UNC-60A and UNC-60B mRNAs were almost equally detected ( Figure 6A , lane 3 ) , while the latter was hardly detectable in sup-12 mutant ( lane 4 ) , consistent with the result with the reporter ( Figure 2H ) and our previous study [30] . To analyze processing patterns of UNC-60B RNAs in body wall muscles , we amplified partially spliced RNAs carrying intron 2B , 3B or 4B with a forward primer in exon 1 and intronic reverse primers ( Figure 6B ) . Partially spliced RNAs committed to UNC-60B , in which exon 1 was connected to exon 2B , were detected in the wild type ( all panels , lane 3 , bands 2 and 3 ) but were undetected in sup-12 mutant ( lane 4 ) , consistent with the result shown in Figure 6A . These results indicated that SUP-12 is required for proper splicing between exon 1 and exon 2B in muscles . In sup-12 mutant , all the introns , including intron 1A , were excised in the only detected RNAs ( Figure 6B , all panels , lane 4 , band 1 ) , while in the wild type , intron 1A is retained in the longest detected RNAs ( all panels , lane 3 , band 1 ) , indicating that SUP-12 represses excision of intron 1A . We next analyzed partially spliced RNAs from the UNC-60A region ( Figure 6C , 6D ) . Although the detected RNAs derived from this region were mixture of those in muscles and in non-muscle tissues , we assumed that differences in their relative amounts could be attributed to functions of SUP-12 in muscles . With a forward primer in intron 1A and a reverse primer in exon 5A ( Figure 6C ) , we detected eight RNA species in sup-12 mutant ( lane 4 , bands 1–6 ) . These RNAs were all the theoretical intermediates in the UNC-60A processing . In the wild type ( lane 3 ) , two of the RNAs ( bands 3 and 6 ) predominated , suggesting that SUP-12 represses their production . In these RNAs , intron 1A alone ( band 6 ) or introns 1A and 2A were retained ( band 3 ) , supporting the idea that SUP-12 represses excision of intron 1A , and weakly of intron 2A , even after introns 3A and 4A are excised . We then analyzed the partially spliced RNAs with the forward primer in exon 1 and intronic reverse primers in introns 2A , 3A and 4A ( Figure 6D ) . All the two ( top panel , band 1–2 ) , four ( middle panel , bands 1–4 ) and eight ( bottom panel , bands 1–7 ) theoretical intermediate RNA species were detected in sup-12 mutant ( lane 4 ) , and relative amounts of the partially spliced RNAs to the pre-mRNAs ( band 1 ) in the wild type ( lane 3 ) and sup-12 mutant ( lane 4 ) were in good accordance with the idea that excision of introns 1A and 2A is facilitated in the absence of SUP-12 . All these analyses of the partially spliced RNAs supported the model that SUP-12 represses excision of intron 1A to preserve exon 1 until exon 2B is transcribed in muscles .
In this study , we have provided genetic and biochemical analyses of the mechanisms for regulation of the muscle-specific alternative processing of the unc-60 pre-mRNA . Figure 7 illustrates models of the pre-mRNA processing deduced from this study . In non-muscle tissues ( Figure 7A ) , intron 1A and the other introns are excised during or after transcription and UNC-60A mRNA is generated . The order of intron removal is not strictly regulated as suggested by the presence of all the theoretical partially spliced RNAs ( Figure 6C , 6D ) . In muscles ( Figure 7B ) , ASD-2b and SUP-12 cooperatively bind to CUAAC repeats and UGUGUG stretch , respectively , in intron 1A to repress excision of intron 1A and weakly of intron 2A during transcription of the UNC-60A region . When UNC-60B-specific region is being transcribed , exon 1 is readily spliced to exon 2B , and introns 3B and 4B are also readily removed in the order of transcription ( Figure 6B ) . Introns 3A and 4A are properly and rapidly excised during the UNC-60B processing ( Figure 6C ) likely due to their small sizes ( 53 nt and 60 nt , respectively ) . This may explain why exon 1 is not aberrantly spliced to exons 3A or 4A but is exclusively spliced to exon 2B to form UNC-60B mRNA . Regulation of tissue-specific alternative polyadenylation may also be involved in the fate-decision of the unc-60 transcript , although the results demonstrated above did not provide conclusive evidence that ASD-2 and/or SUP-12 regulate muscle-specific repression of the polyadenylation site for UNC-60A mRNA . We have demonstrated that ASD-2 and SUP-12 cooperatively represses the 3′-splice site and not the 5′-splice site of intron 1A . Although C . elegans does not have a recognizable branch point consensus or a polypyrimidine tract [42] , a putative branch site for intron 1A is the A at position -19 , between CUAAC repeats and UGUGUG stretch ( Figure 3A ) . This A is the first A upstream from the 3′ splice site and is close to the positions where the putative branch site A is frequently found [43] . It is therefore reasonable to suggest that formation of ASD-2b/SUP-12/RNA ternary complex sterically hinders U2 snRNP auxiliary factor ( U2AF ) bound to the 3′-splice site from recruiting U2 snRNP to the branch site . The situation is quite similar to muscle-specific repression of egl-15 exon 5B , where the Fox-1 family proteins and SUP-12 cooperatively bind to juxtaposed cis-elements overlapping a putative branch site [20] , [38] . Recent microarray analyses of alternatively spliced exons in splicing factor mutants including sup-12 identified many other splicing events affected by multiple splicing factors [44] . Combinatorial regulation by multiple splicing factors may be the common feature in tissue-specific alternative pre-mRNA processing in C . elegans . ASD-2 ortholog in Drosophila , Held out wings ( How ) [45] , [46] , [47] , and that in zebrafish , Quaking A ( QkA ) [48] , are known to be required for muscle development or activity by mutant analyses . Vertebrate orthologs of SUP-12 , known as SEB-4 or RBM24 , are also expressed in muscle tissues and have recently been shown to be involved in myogenic differentiation by knockdown experiments [49] , [50] , [51] , [52] , [53] . However , the target events that these orthologs regulate in muscles remain almost unclear . Considering the highly conserved amino acid sequences and their expression patterns , it is likely that the orthologs of ASD-2 and SUP-12 regulate alternative pre-mRNA processing to produce muscle-specific protein isoforms in higher organisms . In this study , we have presented a model of complex alternative pre-mRNA processing of a gene generating two almost distinct mRNAs . An important aspect of this study is the successful application of a dichromatic fluorescence reporter system to analyze the complex alternative pre-mRNA processing . The asymmetric pair of fluorescence reporter minigenes utilized in this study offers an alternative option for visualizing complex processing patterns besides symmetric pairs of minigenes applied to mutually exclusive exons and cassette exons [32] , [33] . Another example of evolutionarily conserved genes with a structure similar to the unc-60 gene is the cholinergic gene locus; genes encoding choline acetyltransferase ( ChAT ) and vesicular acetylcholine transporter ( VAChT ) share the common first exon , and the other exon ( s ) for VAChT reside in the first intron of the ChAT gene in mammals [54] , Drosophila [55] and C . elegans [56] . The regulation mechanisms presented here would provide insight into the regulation of this kind of genes . We demonstrated that ectopically expressed UNC-60A can compensate for the function of UNC-60B in sarcomeric actin organization in body wall muscles of unc-60B mutant . However , both UNC-60A and UNC-60B have characteristic actin-regulatory activities of ADF/cofilin in vitro with some quantitative differences [27] , [28] , [29]; UNC-60A has strong actin-monomer sequestering and only weak actin-filament severing activities , while UNC-60B has no actin-monomer sequestering and strong actin-filament severing activities . Although UNC-60A can compensate for the function of UNC-60B in body wall muscles , sarcomeric actin filaments in UNC-60A-complemented unc-60B mutant muscles still exhibit minor disorganization ( unpublished data ) , suggesting that UNC-60B is a more suitable isoform . On the other hand , UNC-60B cannot compensate for the function of UNC-60A in the gonadal myoepithelial sheath [29] . This work and our previous works demonstrated that UNC-60A and UNC-60B are specifically adapted for functions in non-muscle and muscle cells , respectively , emphasizing that precise expression of appropriate ADF/cofilin isoforms , unravelled in this study , is important for development of tissue-specific actin-cytoskeletal structures [26] , [29] .
To construct the unc-60E1-E2A-RFP and unc-60E1-E3B-GFP cassettes , unc-60 genomic fragments spanning from exon 1 through 2A and exon 1 through 3B , respectively , were amplified from N2 genomic DNA and cloned into Gateway Entry vectors ( Invitrogen ) carrying either mRFP1 [57] or EGFP ( Clontech ) cDNA by using In-Fusion system ( BD Biosciences ) . M1 and M2 mutations were introduced by mutagenesis with Quickchange II ( Stratagene ) . Expression vectors were constructed by homologous recombination between the Entry vectors and Destination vectors [31] , [33] with LR Clonase II ( Invitrogen ) . Sequences of the primers used in plasmid construction are available in Table S1 . Worms were cultured following standard methods . Transgenic lines were prepared essentially as described [33] using lin-15 ( n765 ) as a host or pmyo-2-mRFP as a marker . Integrant lines were generated by ultraviolet light irradiation as described previously [33] , [58] . Images of fluorescence reporter worms were captured using a fluorescence stereoscope ( MZ16FA , Leica ) with a dual and-pass filter GFP/DsRed equipped with a colour , cooled CCD camera ( DP71 , Olympus ) or a confocal microscope ( Fluoview FV500 , Olympus ) and processed with Metamorph ( Molecular Devices ) or Photoshop ( Adobe ) . RNAi experiments by feeding were performed essentially as described [59] . Briefly , L4 hermaphrodites were transferred to agar plates seeded with bacteria expressing dsRNAs of target genes and their progeny were scored for colour and behavioural phenotypes or used for staining . For RNAi experiment by micro-injection , sense and anti-sense asd-2 RNAs were prepared as described preciously [32] and were annealed at room temperature and 1–5 µg/µl dsRNA was injected into the gonad of young adult hermaphrodites . Injected worms were cultured at 20°C and the colour phenotype of their progeny was evaluated . Total RNAs were extracted from worms by using RNeasy Mini kit ( Qiagen ) and DNase I ( Promega ) . RNAs ( 300–500 ng ) were reverse transcribed using random hexamers and Superscript II ( Invitrogen ) according to manufacturer's protocol . PCR was performed essentially as described previously [31] , [33] . For amplification of partially spliced RNAs , total RNAs were reverse transcribed with PrimeScript II and random hexamers ( Takara ) , and amplified with BIOTAQ ( Bioline ) and analyzed by using BioAnalyzer ( Agilent ) . Sequences of the RT-PCR products were confirmed either by direct sequencing or by cloning and sequencing . Sequences of the primers used in the RT-PCR assays are available in Table S2 . Denatured His-tagged full-length ASD-2b for immunization was purified from denatured bacterial lysate by using Ni-NTA agarose ( QIAGEN ) . Cold-shock inducible expression vectors for His-GST-fused full-length ASD-2b and mRFP1 and FLAG-tagged full-length SUP-12 were constructed by using Destination vectors pDEST-Cold-GST and pDEST-Cold-FLAG ( H . K . ) , respectively . GST-ASD-2b and FLAG-SUP-12 were purified by using Glutathione Sepharose 4B ( GE Healthcare ) and Anti-FLAG M2 Magnetic Beads ( Sigma ) , respectively , and dialyzed against RNA binding buffer ( see below ) . Purified proteins were separated by standard SDS-PAGE and stained with SimplyBlue SafeStain ( Invitrogen ) . Rabbit polyclonal anti-ASD-2b antiserum was generated with denatured recombinant His-ASD-2b protein by Operon Biotechnologies ( Tokyo , Japan ) . IgG fraction ( TD0135-02 ) was prepared from the antiserum by Medical & Biological Laboratories ( Nagoya , Japan ) . Worm lysates were extracted from synchronized L1 larvae , separated by neutral polyacrylamide gel electrophoresis ( NuPAGE , Invitrogen ) and transferred to nitrocellulose membrane ( Protran BA85 , Whatman ) . Western blotting was performed with 15 µg/ml anti-ASD-2b ( TD0135-02 ) or 1∶40 , 000-diluted anti-actin monoclonal antibody ( Ab-1 , Calbiochem ) and 1∶1 , 000-diluted HRP-conjugated anti-rabbit IgG antibody ( Pierce ) or 1∶10 , 000-diluted HRP-conjugated anti-mouse IgM antibody ( Calbiochem ) . Chemiluminescence signals ( West Dura , Thermo ) were detected by using LAS4000 ( GE Healthcare ) . For staining with anti-ASD-2b , mixed stages of N2 and asd-2 ( yb1540 ) worms were fixed with Bouin's fixative ( 15∶5∶1 mixture of saturated picric acid , formalin and acetic acid ) supplemented with 25% methanol and 1 . 25% 2-mercaptoethanol for 60 min at room temperature , washed with phosphate-buffered saline ( PBS ) and permeabilized with 5% 2-mercaptoethanol and 1% Triton X-100 in PBS at 37°C for 30 hours . Fixed worms were treated with blocking buffer ( 0 . 5% skim milk and 0 . 5% bovine serum albumin ( BSA ) in PBS ) for 2 hours at room temperature and stained with 6 µg/ml anti-ASD-2b ( TD0135-02 ) as a primary antibody in blocking buffer for 24 hours at room temperature and then with 2 µg/ml Alexa488-conjugated goat anti-rabbit IgG ( Invitrogen ) as a secondary antibody together with 1 µg/ml Hoechst 33258 ( Hoechst ) in blocking buffer for 2 hours at room temperature . Fluorescence images were captured by using a compound microscope ( DM6000B , Leica ) equipped with a colour , cooled CCD camera ( DFC310FX , Leica ) or an inverted fluorescence microscope ( Nikon TE2000 ) equipped with a monochrome CCD camera ( SPOT RT , Diagnostic Instruments , Inc ) . Staining with anti-UNC-60A and anti-MyoA were performed as described previously [30] . Actin filaments were visualized by staining with tetramethylrhodamine-phalloidin as described previously [60] . 32P-labelled RNA probes were generated by in vitro transcription with [α32P] UTP ( Perkin Elmer ) and T7 RNA polymerase ( Takara ) . Sequences of template oligo DNAs are available in Table S3 . Gel-purified RNA probes alone or with increasing amounts of recombinant protein ( s ) were incubated in 25 µl of RNA binding buffer ( 20 mM HEPES-KOH ( pH7 . 9 ) , 150 mM KCl , 5% glycerol , 1% Triton X-100 , 1 mM DTT and 0 . 1 mM PMSF ) supplemented with 100 ng/µl E . coli tRNA and 50 ng/µl BSA for 30 min at 20°C . Each sample was separated on a non-denaturing 4% polyacrylamide gel and analyzed with a fluoro-imaging analyzer ( FLA-3000G , Fuji Film ) . His-GST-fused recombinant full-length ASD-2b and mRFP1 proteins were immobilized on glutathione sepharose 4B beads ( GE Healthcare ) and incubated with His-SUP-12 in 100 µl of pull-down buffer ( 20 mM HEPES-KOH ( pH7 . 9 ) , 150 mM KCl , 1% Triton X-100 , 1 mM DTT and 0 . 1 mM PMSF ) supplemented with 100 ng/µl E . coli tRNA , 50 ng/µl BSA and 0 , 30 , 100 , or 300 nM of unc-60-I1A RNAs ( Operon Biotechnologies ) for 30 min at 20°C . The sequences of the unc-60-I1A RNAs: unc-60-I1A-WT , 5′-UUUUUGCCUAACCUAACCUAACCUAUGUGUGCCUGUUUU-3′; unc-60-I1A-M1 , 5′-UUUUUGCCAAACCAAACCAAACCUAUGUGUGCCUGUUUU-3′; unc-60-I1-M2 , 5′-UUUUUGCCUAACCUAACCUAACCUAUAUAUACCUGUUUU-3′ . Beads were washed four times with 1 ml pull-down buffer . Bound proteins were eluted with LDS sample buffer and separated by NuPAGE ( Invitrogen ) . Gels were stained with SimplyBlue SafeStain ( Invitrogen ) and detected and analyzed by using LAS4000 ( GE Healthcare ) . | Muscle is a specialized organ with specialized contractile apparatuses . A number of genes encoding contractile apparatus-related proteins undergo muscle-specific pre–mRNA processing . However , the molecular mechanisms and consequences of muscle-specific alternative pre–mRNA processing remain largely unknown . In this study , we reveal regulation mechanisms of pre–mRNA processing of the unc-60 gene locus , encoding two tissue-specific isoforms of ADF/cofilin in C . elegans . The unc-60A and unc-60B genes share only the first exon , and UNC-60B protein is specifically expressed in muscle . We visualize the tissue-specific processing patterns of the unc-60 pre–mRNA with green and red fluorescent proteins in living worms . We provide genetic , biochemical , and immunohistochemical evidence that muscle-specific RNA–binding proteins ASD-2 and SUP-12 cooperatively bind to specific motifs in intron 1A to retain intron 1A , which leads to skipping of exon 2A through 5A and splicing between exon 1 and 2B . Consistently , disruption of the splicing factors leads to expression of UNC-60A in muscle and suppresses paralysis of an unc-60B-specific mutant . Our study raises a model of step-by-step execution of complex co-transcriptional pre–mRNA processing and provides insight into the fate decision of the entire transcript . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"rna",
"interference",
"gene",
"function",
"gene",
"expression",
"muscle",
"fibers",
"biology",
"molecular",
"biology",
"rna",
"rna",
"processing",
"cell",
"biology",
"nucleic",
"acids",
"genetic",
"screens",
"genetics",
"cellular",
"types",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Muscle-Specific Splicing Factors ASD-2 and SUP-12 Cooperatively Switch Alternative Pre-mRNA Processing Patterns of the ADF/Cofilin Gene in Caenorhabditis elegans |
The dendritic tree contributes significantly to the elementary computations a neuron performs while converting its synaptic inputs into action potential output . Traditionally , these computations have been characterized as both temporally and spatially localized . Under this localist account , neurons compute near-instantaneous mappings from their current input to their current output , brought about by somatic summation of dendritic contributions that are generated in functionally segregated compartments . However , recent evidence about the presence of oscillations in dendrites suggests a qualitatively different mode of operation: the instantaneous phase of such oscillations can depend on a long history of inputs , and under appropriate conditions , even dendritic oscillators that are remote may interact through synchronization . Here , we develop a mathematical framework to analyze the interactions of local dendritic oscillations and the way these interactions influence single cell computations . Combining weakly coupled oscillator methods with cable theoretic arguments , we derive phase-locking states for multiple oscillating dendritic compartments . We characterize how the phase-locking properties depend on key parameters of the oscillating dendrite: the electrotonic properties of the ( active ) dendritic segment , and the intrinsic properties of the dendritic oscillators . As a direct consequence , we show how input to the dendrites can modulate phase-locking behavior and hence global dendritic coherence . In turn , dendritic coherence is able to gate the integration and propagation of synaptic signals to the soma , ultimately leading to an effective control of somatic spike generation . Our results suggest that dendritic oscillations enable the dendritic tree to operate on more global temporal and spatial scales than previously thought; notably that local dendritic activity may be a mechanism for generating on-going whole-cell voltage oscillations .
The dendritic tree contributes significantly to the elementary computations a neuron can perform , both by its intricate morphology and its composition of voltage-gated ionic conductances [1] . Such active conductances can underlie a wide variety of dynamical behaviors , amongst others dendritic spikes and ongoing oscillations of the dendritic membrane potential [2] . Such active dendritic phenomena have been suggested as mechanisms endowing single neurons with significant computational power [3] and flexibility in the way the dendritic tree processes its inputs: whether as a global element , effectively collapsing the tree into a single functional compartment or with various parts of the tree acting as independent processing elements [4] , [5] . While the possibility of powerful and flexible dendritic processing has been of great interest , the precise conditions under which dendrites can act independently or globally remain largely to be determined . In this report we address this key question , focusing specifically on the case where active properties lead to sustained intrinsic membrane potential oscillations in the dendrites . We develop a theoretical formalism , allowing for a succinct yet powerful description of the dendritic tree dynamics and yielding conditions under which the tree acts as a global oscillatory unit and how such action in turn controls spiking responses of the neuron . Membrane potential oscillations have been demonstrated in various types of neurons . Prominent intrinsic subthreshold oscillations have been found in stellate cells from entorhinal cortex layer 2 [6] , [7] , neurons from the frontal cortex [8] , neurons from the amygdala complex [9] , [10] , and pyramidal cells and interneurons from the hippocampal CA1 area [11] , [12] . Although these membrane potential oscillations are normally recorded at the soma and thus are considered to be of somatic origin , several lines of evidence suggest dendritic loci of generation . First , many of the conductances thought to underlie the generation of such oscillations reside predominantly in the dendrites , sometimes specifically in the distal parts of the dendritic tree . For example , in the apical dendrites of hippocampal CA1 pyramidal neurons , the density of increases strongly with distance from the soma [13] , and reaches very high values in the thin distal branches [14] . Second , several studies have suggested the existence of clusters of ionic conductances that are responsible for the generation of dendritic spikes [15] . While most of the direct electrophysiological evidence regards excitable behavior , demonstrating the generation of dendritic spikes in response to sufficient levels of depolarization , mathematical analysis has shown that neural membranes exhibiting excitability can readily pass to oscillatory regimes in an input-dependent manner ( e . g . see [16] ) . Third , in several cases , oscillations have been directly recorded in dendrites . For example , recordings from hippocampal CA1 pyramidal neurons have demonstrated ongoing oscillations in the dendrites that include repetitive dendritic spikes , presumably involving Ca currents [17] . Furthermore , significant intrinsic dendritic oscillations have been observed in several neuronal preparations that depended on the interplay between the non-linear properties of NMDA synaptic receptors and intrinsic voltage-dependent currents [18] , [19] . Crucially , while the onset of these oscillations was conditional on the activation of the NMDA synapses , the oscillations themselves were produced by mechanisms that were intrinsic to the postsynaptic cell and not by periodically structured synaptic inputs . Since NMDA receptors are largely localized on dendritic spines , and are hence electrotonically removed from the soma , these data may also argue for a non-uniform and local dendritic generation of membrane potential oscillations . Taken together , these experimental results suggest that dendritic trees can function as oscillators , perhaps conditional on the level of background depolarization or the presence of neuromodulators [20] , while leaving open the question whether global cell-wide voltage oscillations could result from local dendritic mechanisms that are intrinsic even to distal dendrites and hence only weakly coupled to the soma electrotonically . Indeed , multiple intrinsic dendritic oscillators have been proposed to underlie the recently discovered intricate firing pattern of entorhinal grid cells [21]–[23] . This influential model suggests that the functional responses of entorhinal neurons recorded in behaving animals are a direct consequence of the generation of independent oscillations that are intrinsic to individual dendrites . Hence , this model presupposes the existence of multiple oscillators that are integrated at the soma , leading to the questions of how such dendritic oscillators may interact with the soma and with each other , and what sorts of collective behaviors the electrotonic structure of the dendritic tree might impose on the oscillations . In this paper , we study the dynamics of such interacting oscillators and their impact on signal propagation in single neurons , using mathematical analysis corroborated by numerical simulations of biophysical models . We treat the dendritic tree of a neuron as a network of oscillators coupled by stretches of relatively less active cable . This prompts us to combine two analytical methods: weakly coupled oscillator theory and cable theory . The theory of weakly coupled oscillators has been extensively used previously to study synchronization of multiple oscillators residing in separate cells interacting through synapses or gap junctions [24] . Since we focus on intradendritic oscillators which are continuously coupled via the membrane voltage , we use cable theory [25] to compute their interactions . We find that intradendritic oscillations can exhibit complex patterns of phase-locking . We characterize how this phase-locking depends on the intrinsic properties of the oscillators and on the membrane properties of the segment connecting them . Finally , we demonstrate how input to the dendritic oscillators can control the phase-locking and how in turn the phase-locked configuration can control somatic spike generation . These results provide a rigorous mathematical framework for the study of interacting dendritic oscillations that can be applied in the future to specific systems of interest , and also point to ways in which such oscillations can be utilized for non-trivial single cell computations .
The basic behavior of the system can be most easily understood by examining a simplified situation where the oscillators have a phase response curve that is approximately sinusoid and the perturbations from the cable are also nearly sinusoidal ( e . g . when the oscillators are subthreshold with simple sinusoidal voltage traces ) . Hence the first Fourier component dominates in both and . The interaction function is then ( 6 ) where is a positive coefficient characterizing the strength of the coupling ( see Equation 22 in Methods ) . The term gives the effective phase delay in the interaction between the two oscillators ( figure 1Aii ) . In this term depends on the properties of the oscillators and summarizes the effect of cable filtering . It depends on the properties of the dendritic cable: , , and ( see Methods ) . Using Equation 5 it is easy to show that the evolution of the phase difference between two identical oscillators is given by ( 7 ) The fixed points are the in-phase solution and the anti-phase solution ( figure 1Aiii ) . The stable phase-locked solutions are those fixed points where the derivative of Equation 7 with respect to is negative: ( 8 ) The synchronous solution is thus stable when . When this solution is stable the anti-phase solution is unstable and vice versa . Notice that if we fix the properties of the oscillators , the constant is fixed . Then the value of uniquely determines which is the stable state ( figure 1Aiv ) . Hence , to understand how the dendrite behaves as a function of the key properties of the cable we need only to look at how these affect . In the next sections we describe the behavior of with the consequent effect on phase-locking . The explicit expressions for the scaling of with the various parameters considered below are given in the Methods . In the previous section we limited our description and analysis to oscillators with a nearly sinusoidal phase response curve that receive perturbations which are also sinusoidal . In this way we could demonstrate how the parameters that define the oscillator and cable properties affect the phase-locking behavior of the system . However , as consequence , we only obtained and analyzed symmetric interaction functions . For such coupling functions , only the in-phase and anti-phase solutions are possible of which one is stable and one unstable . When and cannot be well approximated by a single Fourier component we need to take into account higher order terms . Including more Fourier components is likely to lead to asymmetry or skew of and , as we will show next , this affects the possible phase-locking behaviors . Above we developed a framework for analyzing the behavior of local oscillators embedded in the dendritic tree . Now we turn to the question of how such oscillating dendrites respond to inputs and impact the output of the neuron . We will show that the external synaptic input can control the phase-locked configuration of the dendritic oscillators and that this phase-locked configuration can then be transmitted through patterning of the cell's action potentials . While a thorough analysis is beyond the scope of the present study , we give several salient illustrative examples using a model with a branched oscillating dendritic tree and a spike-generating soma . More specifically the model consists of a passive branching dendritic compartment with two Morris-Lecar type II oscillators at its two distal ends and an excitable soma that , for simplicity , we describe with an integrate and fire mechanism ( figure 7A ) . Above we showed that under certain conditions , depending on the skew of the interaction function , the dendritic tree can be in a phase-locking regime where two stable phase-locked states co-exist ( see figures 3C and 4C ) . In such a bistable regime , well-timed inputs to one or more dendritic oscillators can switch the locking between in-phase and anti-phase . Clearly , the membrane potential fluctuations at the soma depend on whether the dendritic oscillators are synchronized or not . In our model , they are largest in amplitude when the dendritic oscillators are in-phase . The soma can show this difference with its spiking pattern when such large amplitude fluctuations are supra-threshold , while smaller fluctuations ( e . g . with asynchronous oscillators ) are not . In figure 7 we illustrate the above mechanism . The initial parameters are such that both the in-phase and anti-phase state of the dendritic oscillators are stable ( black dotted line in figure 7C ) . Oscillators starting from an initial phase difference move into the synchronous phase-locked state ( red curve in figure 7B ) . This consequently leads to repetitive somatic spiking ( blue traces in middle and bottom panel ) . A brief depolarizing current pulse to one of the oscillators ( see black trace in top panel of figure 7B ) moves them into the anti-synchronous state and the somatic spiking ceases . A subsequent synchronous current pulse to both dendritic oscillators can switch them back into the synchronous state and hence restart the spiking . Note that all the stimuli here are excitatory , yet depending on their timing , they can have a net excitatory or inhibitory effects on the cell's spiking . We have also hinted , in a previous section , at another mechanism by which inputs to the dendrites can affect the phase-locked state . The input amplitude can change the oscillator frequency which in turn has an effect on the stability of the phase-locked state ( see figure 1D ) . In figure 7B at time sec we increase the amplitude of the current input impinging on the oscillators which causes the system to move out of the bistable regime . The synchronized state loses stability and the oscillators gradually move into anti-phase locking . As a result , the soma stops spiking ( at time sec ) . Note that the electrotonic separation between the oscillators remains constant ( black dotted line in figure 7D ) but that the bifurcation diagram itself changes . In turn , a decrease in the excitatory input would reinstate spiking . Hence , this mechanism allows the cell to encode an inverse of the input amplitude , or the inverse of the excitatory input rate .
The question of how local cellular processes may lead to global behavior has been of great interest for some time , in particular with respect to the signal propagation in extended structures such as the dendritic trees of cortical neurons . One of the aspects that remains a subject of active debate , is the dendritic mechanisms that ensure that local inputs on the dendrites – and in particular on the distal dendrites – have an impact on the global signal processing in the cell and ultimately on spike generation . We addressed this key question focusing specifically on the case of oscillatory dendrites . Thus , we studied the dynamics of dendrites that show intrinsic oscillations due to active voltage-dependent currents that present strong spatial inhomogeneities , hence leading to discrete oscillatory segments . Our prime question was to understand how global dendritic behavior , in this case the phase-locked oscillations , can arise from interactions between such local oscillators . To do so we developed an analytical framework to describe and understand the behavior of interacting dendritic oscillators and their impact on signal propagation within a neuron . Our goal was to understand when the oscillators within the dendrite would lock and hence the whole dendritic tree would act as a single oscillatory unit . Using the weakly coupled oscillator framework we have identified the requirements for the various phase-locking regimes of the dendritic oscillators . We characterized how the type of phase-locking depends on the intrinsic properties of the oscillators as well as on the membrane properties of the dendrite segment connecting them . We find that a central parameter in determining the phase-locked solutions is the electrotonic distance between the oscillators . This distance determines how strongly the dendritic cable filters the interactions between the oscillators , thereby determining the delay between the interactions . As a function of the electrotonic distance the phase-locking of identical oscillators alternates between in-phase or synchronized solutions and anti-phase solutions . We also showed how the phase-locking is affected by the presence of voltage-dependent conductances in the cable that connects the oscillators . Using the quasi-active approximation of the cable [29] , [30] we found that the dependence of the stable phase-locked solution on the electrotonic distance is typically amplified by regenerative conductances ( i . e . ionic conductances that amplify a voltage perturbation ) , whereas it is counteracted by restorative conductances ( i . e . ionic conductances that counteract voltage perturbations ) ( see also [28] ) . It should be noted that the linearization of the active conductances in the dendrites is appropriate for small amplitude oscillations in the dendrite and is therefore in general a better approximation for subthreshold oscillations than for spiking oscillators . The mathematical approach that we used , builds on several studies which focused on the interaction between two neurons with repetitively spiking somata that interact via inputs at the dendrites [26]–[28] . A crucial difference with these studies is that rather than coupling via discrete synaptic events , we treat continuous coupling between the oscillators via the current-conducting cables . One consequence of the continuous coupling is that one needs both the phase response function and the voltage trajectory of the oscillators in order to compute the interaction functions and ultimately the phase-locked solutions . By computing the convolution of the voltage trajectory and the phase response function , which yields the interaction function for directly coupled oscillators , it is possible to get some insight into the types of phase-locked solutions that can be expected . The skew of the interaction function can show whether regimes can be expected in which both in-phase and anti-phase solutions are stable . Both the voltage trajectory of an oscillator and its phase response function can be determined numerically from a model of an oscillator and , at least in principle , also experimentally ( see , for example , [34] ) . In the final section of our study we demonstrated how inputs to the dendritic tree can set the phase-locked state and how in turn the phase-locked configuration can control somatic spike generation . The first can for instance be accomplished by changing the frequency of the oscillators with the external input . The soma can subsequently detect the amplitude of the membrane potential fluctuations since this is affected by the phase-locked configuration . The time scale at which the dendritic oscillators move from one solution to another is set by the strength of the interactions between the oscillators . This time scale can be much longer than that of the different components of the system , e . g . the membrane time constant or the period of the oscillators . In this way , the phase difference between the oscillators can function as a memory . Related ideas have been previously discussed by Huhn et al [35] . We also showed that in the bistable phase-locked regime the state of the dendrites is easily set by transient inputs and “read-out” by the soma . This also can endow the neuron with a memory since brief external inputs can switch the neuron from a spiking to a quiescent mode and vice versa . Interestingly we showed that both the turn-on and turn-off signals ( inputs ) can be excitatory , their final effects defined by their timing . The focus of our report is complementary to that of a recent theoretical study of the subthreshold oscillations in the dendrites of mesencephalic dopaminergic neurons [36] . As these cells do not show any indication of distinct dendritic oscillators , the whole cell was modeled as one continuous oscillator with gradients in oscillator properties along the dendrites . Moreover , since there were no distinct oscillators , in their analysis Medvedev and colleagues assumed strong voltage coupling between neighboring compartments , enforcing synchronized oscillations throughout the cell . In contrast , our approach assumed weak coupling between the dendritic oscillators . This would not be appropriate for a spatially continuous oscillator . However , it is not possible to state in general at what precise electrotonic distance between two oscillators the weak coupling assumption becomes valid , since it depends on the strength of the interaction currents with respect to the intrinsic currents of the oscillators . However , our numerical simulations for a dendritic cable without the assumption of weak coupling , show that the phase-locking behavior of Morris-Lecar oscillators is consistent with weak coupling . One of the aims of the present paper was to set up an analytical framework for studying interacting dendritic oscillators . This opens up a wide range of questions that were outside the scope of the present study . For example , we focused our analysis on identical oscillators , while it is likely that dendritic oscillators will vary in their properties throughout the dendritic tree . For example , the diameter of the dendrites , which typically becomes smaller with increasing distance from the soma , can affect the intrinsic frequency of the oscillators . A gradient in the frequency of distinct oscillators is likely to lead to more complex phenomena such as traveling waves ( see , for example , [37] ) . In fact the major focus of our study is to explore how local dendritic mechanisms may lead to oscillations expressed globally in the cell and hence visible at the soma , for example in somatic intracellular recordings . Our analysis showed that even electrotonically far removed dendritic oscillators can lead to voltage oscillations that significantly affect the soma voltage and hence spike generation . This suggests several experimentally testable predictions . In one possible experiment one can take advantage of imperfect space clamp in a electrotonically extended neuron . As a proof of principle , in a neuron where the oscillations are generated distally in the dendritic tree , voltage clamping the soma would not block such oscillations , and these should be seen in the current necessary to hold the somatic potential . In fact , results from [18] point in this direction , where in chick spinal cord neuron NMDA-dependent intrinsic oscillations were not blocked by somatic voltage clamp . A further prediction stems from the weak coupling between active dendrites . If active oscillations , such as periodically generated dendritic spikes , are generated in different segments of the dendritic tree , our analysis predicts that such spikes should interact and should exist in a stable phase-locked configuration , e . g . synchrony . Hence , should one of the dendritic segments be phase-shifted , such perturbation should 1 . propagate to the other segment ( the other segment should be phase reset ) 2 . the dendritic spikes should return to the phase-locked configuration 3 . the time scale of this return should be relatively long and determined by the electrotonic distance between the active segments . While difficult such experiments are possible using the multiple dendritic recording techniques , such as those developed by Davie et al [38] in Purkinje cells . A recent model for the grid field properties of the entorhinal cortex layer II stellate cells [21] , [22] , [39] relies precisely on the ingredients considered in the present study . The model assumes that different dendritic branches emanating from the soma of these cells function as distinct oscillators . The oscillations are modulated by external inputs and the interference of the oscillators eventually determines the somatic spiking . Crucially , the model assumes that the dendritic oscillators operate independently . At a first glance , our results appear to argue against this: the various oscillators should phase-lock ( hence lose their independence ) even when the mutual coupling is weak . However , in principle , the locking may be slower than the behavioral time scale , allowing the oscillators to act quasi-independently on the behavioral time scale . Our analysis provides the appropriate framework to examine these issues: the scaling of locking in time and the biophysical implementation of grid-field formation via dendritic oscillators . Above we studied relatively simple cell geometries , however these form basic building blocks for more complex dendritic trees . Thus our framework should be valid for understanding global voltage oscillations in more realistic models of spatially extended cells . We would like to emphasize at this point that our general framework should also hold when – in addition to the distinct oscillators distributed throughout the dendritic tree – also the soma is regarded as an oscillator . These and other issues will be addressed in future publications . The framework we have developed , builds on the extensive mathematical theory of coupled oscillators and nestles nicely below the complexity of full compartmental models of neuronal dendritic trees . Yet our framework is sufficiently powerful and clear to both take into account certain key aspects of the dendritic tree structure and to be amenable to theoretical analysis of the dynamics of active dendrites and the computational function of such dendritic structures . These remain an active focus for further investigations .
We analyze the behavior of a system of two oscillators that are coupled via a cable . For this we need to compute the interaction between the two oscillators . Our approach is as follows . The oscillators provide the periodically forced end conditions for the cable equation . Assuming weak coupling the phase difference between the oscillators does not change significantly within one period of the oscillation . Thus we can solve the cable equation with such boundary conditions and leave the phase difference as a free parameter . In turn , the solution of the cable equation yields the currents flowing into and thereby perturbing the two oscillators at its ends . We let denote the membrane potential ( in millivolts ) along the cable at position ( in centimeters ) and at time ( in milliseconds ) . The passive properties of the cable are determined by a membrane time constant ( in milliseconds ) and a length constant ( in centimeters ) . The cable also expresses a voltage-dependent conductance with a gating variable with activation function and time constant ( in milliseconds ) . The equations governing the membrane potential and the gating variable along the cable ( excluding the oscillators ) are ( 9 ) where is the leak reversal potential , is the reversal potential of the active current , and is the ratio of the maximal conductance of the active current to the leak conductance . The two oscillators form the periodically forced end conditions of the cable: ( 10 ) with and being the voltage traces of the two oscillators A and B that evolve according to ( 11 ) where is the membrane capacitance ( in F/cm ) , is the leak conductance ( in mS/cm ) , summarizes the voltage-dependent membrane currents generating the oscillations with the vector of gating variables given by standard kinetic equations ( e . g . see Equations 28 and 29 ) . The terms describe the perturbing currents from the cable to each oscillator with the small parameter denoting the coupling . For a cable with diameter ( in cm ) and oscillators with membrane surface area ( in cm ) , , where is the intracellular resistivity of the dendritic cable ( in kcm ) . The functions are given by ( 12 ) Consider identical oscillators when both and are dominated by the first Fourier component . One can show that the interaction function is given by ( 21 ) where is a positive coefficient , is a constant resulting from the cable filtering , is a constant that results from the specific properties of the oscillators and is a constant ( see figure 1A ) . The expressions for the parameters are ( 22 ) ( 23 ) ( 24 ) ( 25 ) where and are , respectively , the absolute value and the angle of the complex number . The shape of the interaction function is determined by Equations 17 , 18 and 20 . When the electrotonic separation between the two oscillators goes to zero , we have a system of directly coupled oscillators and the interaction function reduces to ( 27 ) where the constant . Introducing an electrotonic separation between the oscillators changes the shape of as a result of the cable filtering . When substituting Equation 18 into Equation 20 one sees that the symmetry of can only be affected by the -dependent term involving the voltage trace of oscillator B . As increases , the increasing cable filtering – determined by the absolute value of the term in Equation 18 – leads to dominance of a single Fourier component . Note that it is not necessarily the first Fourier component that will dominate . When a higher order Fourier component can be the dominant one . The equations for the Morris-Lecar type II oscillator [32] with parameters as in [40] read ( 28 ) with F/cm , mS/cm , mS/cm , mS/cm , mV , mV , mV , , A/cm , and where , , and . The equations describing the subthreshold oscillator are of the same form as those used by Morris and Lecar [32] . The oscillatory dynamics emerge from the interaction between the persistent sodium current and the hyperpolarization activated inward current . The current descriptions are based on the data from [33] , [41] . The dynamics of are described by a single gating variable with activation function and time constant ( in milliseconds ) . The voltage-dependent activation of is described by and is instantaneous . The equations read ( 29 ) with F/cm , mS/cm , mS/cm , mS/cm , mV , mV , mV , , A/cm , and where , , and . The numerical simulations for figure 4 , 6 and 7 used Morris-Lecar type II oscillators and simulations for figure 5 used the subthreshold oscillator model described above . The cable was discretized into isopotential compartments with electrotonic length . The perturbing currents from the cable to , for example , oscillator A are of the form with and denoting the membrane potential of the first two compartments . The parameter determines the coupling between the cable and the oscillators and is specified in the different figure captions . Simulations for figure 7 include a soma with an integrate and fire mechanism with a fixed threshold at mV . When the threshold is reached a spike is generated with a 1 ms peak at 30 mV after which the somatic is reset to mV for 4 ms . The phase response curves were calculated by determining the system's adjoint [24] . | A central issue in biology is how local processes yield global consequences . This is especially relevant for neurons since these spatially extended cells process local synaptic inputs to generate global action potential output . The dendritic tree of a neuron , which receives most of the inputs , expresses ion channels that can generate nonlinear dynamics . A prominent phenomenon resulting from such ion channels are voltage oscillations . The distribution of the active membrane channels throughout the cell is often highly non-uniform . This can turn the dendritic tree into a network of sparsely spaced local oscillators . Here we analyze whether local dendritic oscillators can produce cell-wide voltage oscillations . Our mathematical theory shows that indeed even when the dendritic oscillators are weakly coupled , they lock their phases and give global oscillations . We show how the biophysical properties of the dendrites determine the global locking and how it can be controlled by synaptic inputs . As a consequence of global locking , even individual synaptic inputs can affect the timing of action potentials . In fact , dendrites locking in synchrony can lead to sustained firing of the cell . We show that dendritic trees can be bistable , with dendrites locking in either synchrony or asynchrony , which may provide a novel mechanism for single cell-based memory . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neuroscience/neuronal",
"signaling",
"mechanisms",
"neuroscience/neuronal",
"and",
"glial",
"cell",
"biology",
"neuroscience/theoretical",
"neuroscience"
] | 2009 | The Role of Ongoing Dendritic Oscillations in Single-Neuron Dynamics |
Accurate estimates of mutation rates provide critical information to analyze genome evolution and organism fitness . We used whole-genome DNA sequencing , pulse-field gel electrophoresis , and comparative genome hybridization to determine mutation rates in diploid vegetative and meiotic mutation accumulation lines of Saccharomyces cerevisiae . The vegetative lines underwent only mitotic divisions while the meiotic lines underwent a meiotic cycle every ∼20 vegetative divisions . Similar base substitution rates were estimated for both lines . Given our experimental design , these measures indicated that the meiotic mutation rate is within the range of being equal to zero to being 55-fold higher than the vegetative rate . Mutations detected in vegetative lines were all heterozygous while those in meiotic lines were homozygous . A quantitative analysis of intra-tetrad mating events in the meiotic lines showed that inter-spore mating is primarily responsible for rapidly fixing mutations to homozygosity as well as for removing mutations . We did not observe 1–2 nt insertion/deletion ( in-del ) mutations in any of the sequenced lines and only one structural variant in a non-telomeric location was found . However , a large number of structural variations in subtelomeric sequences were seen in both vegetative and meiotic lines that did not affect viability . Our results indicate that the diploid yeast nuclear genome is remarkably stable during the vegetative and meiotic cell cycles and support the hypothesis that peripheral regions of chromosomes are more dynamic than gene-rich central sections where structural rearrangements could be deleterious . This work also provides an improved estimate for the mutational load carried by diploid organisms .
Mutations can arise in genomes as the result of errors that occur during DNA replication , and the repair of DNA lesions [1] , [2] . Mutations such as base substitutions , small insertions and deletions , and large-scale rearrangements are raw materials for adaptive evolution [3]–[5]; however , the deleterious nature of most mutations imposes a fitness cost . In asexual organisms deleterious mutations can accumulate in successive generations . This phenomenon , known as Muller's ratchet , can cause a continuous decrease in fitness and population size in small asexual populations [6]–[8] . In sexual organisms , deleterious mutations can be removed from the population by meiotic recombination and mating [6] , [9] . While this removal of mutations is thought to provide a fitness advantage for sexual organisms , several studies have shown that recombination is itself mutagenic [10]–[12] . Meiosis can also generate new allelic combinations [13] , thus increasing genetic variation and the rate of adaptation to new environments [14] . Therefore , obtaining accurate estimates of mutation rate in vegetative and meiotic cell cycles is important for understanding disease progression , genome evolution , species divergence times and patterns of selection ( reviewed in [15] , [16] ) . These measures also improve our estimates of the mutational load carried by organisms , which are crucial to understand the evolutionary role of sex and recombination . Several genome-wide measurements have been performed to determine the vegetative base substitution rate in a variety of organisms ( reviewed in [16] ) . In baker's yeast , for example , the base substitution rate in haploid mutation accumulation lines grown vegetatively was estimated to be 3 . 3×10−10 substitutions per base per cell division [17] . Importantly , there are no genome wide estimates of the meiotic mutation rates in any organism . However , several lines of correlative and experimental evidence suggest that mutation rates in meiosis are higher than in vegetative growth . First , the phenotypic reversion rates of three independent mutations in S . cerevisiae were observed to be six to twenty-fold higher in meiosis compared to vegetative growth [10] . Second , several studies showed a high mutation rate ( ≥100-fold elevated ) associated with DSB repair of a broken chromosome [11] , [12] , [18] , [19] . Mutations are thought to occur due to error-prone DNA synthesis and/or the absence , or lack of bias , of DNA mismatch repair . Although the mutation rate estimates are for vegetative DSB repair , homologous recombination in meiosis is initiated by the programmed introduction of DSBs [20]–[21] . Lastly , a positive correlation between genetic diversity and meiotic recombination rates has been observed in several organisms [22]–[27] . Curiously , Noor [28] did not see an association between recombination hotspots or DSB sites and sequence divergence between two yeast species ( lack of a correlation , or a negative correlation ) . A concern about most correlation analyses is that they assume that DSB sites are conserved between individuals of the same species and among species . For yeasts , conservation of meiotic DSB sites was recently reported between different species [29] . In this study we used deep DNA sequencing , pulse-field gel electrophoresis ( PFGE ) , and comparative genome hybridization ( CGH ) to determine nuclear mutation rates in vegetative growth and meiosis in diploid mutation accumulation lines of S . cerevisiae . S . cerevisiae is an ideal model organism to obtain such rates because it undergoes rapid vegetative growth ( ∼2 hr cell cycle ) and can complete meiosis in ∼10 hours . Wild isolates of S . cerevisiae are mostly diploid [30] , [31]; importantly , diploid strains can maintain recessive lethal mutations that can comprise 30% to 40% of deleterious mutations [32] , [33] . Vegetative lines were subjected to bottlenecks , from one cell to a colony , every 20 generations , for a total of ∼1740 generations . The meiotic lines underwent 50 meioses and 1 , 000 intervening vegetative generations . While this scheme made it difficult to directly estimate meiotic mutation rates , it was compatible with work indicating that the meiotic cycle is infrequent ( for Saccharomyces paradoxus one meiotic cycle/1 , 000 vegetative cycles; [34] ) . Such a scheme also provides information on how mutations created in the vegetative cycle are propagated as the result of meiosis . As described below , our observations indicate that the baker's yeast diploid genome is highly stable in the vegetative and meiotic cell cycles .
To measure vegetative and meiotic mutation rates in the nuclear genome , we performed mutation accumulation studies in the SK1 homothallic strain of yeast , which grows rapidly in rich media and can complete meiosis in approximately 10 hours [35] . The starting strain for this work , EAY2531 ( relevant genotype MATa/MATalpha , HO/HO ) , is , with exception of the MAT locus , fully homozygous . The spore viabilities of tetrads derived from EAY2531 are greater than 95% . EAY2531 was sequenced using both single and paired end approaches covering 96% of the genome at 64-fold average coverage ( Materials and Methods; Table S1 ) . Data can be retrieved from the European Nucleotide Archive ( http://www . ebi . ac . uk/ena ) using the accession number: ERA007227 . The high sequence coverage allowed us to assemble a high quality reference SK1 genome , accessed in http://steinmetzlab . embl . de/SK1 . Vegetative and meiotic mutation accumulation lines were initiated from EAY2531 ( Materials and Methods; Figure 1 ) . Twenty vegetative lines labeled 1B to 20B were subjected to vegetative growth bottlenecks , from one cell to a colony , every 20 generations for a total of ∼1740 generations ( 87 bottlenecks ) . The twenty meiotic lines labeled 1T to 20T were subject to a bottleneck every meiosis by isolating one complete tetrad that was separately germinated to form colonies . The resulting colony was sporulated and the bottleneck was repeated for 50 meioses and 1 , 000 intervening vegetative generations . At the end of the bottleneck experiments , cells from the final B ( 1B-87 to 20B-87 ) and T ( 1T-50 to 20T-50 ) generations were sporulated and tetrad dissected to assess fitness . As shown in Table S2 , all of the meiotic lines displayed spore viabilities similar to the parental line , indicating that they had not acquired recessive lethal mutations or they had been removed by recombination and mating . Furthermore , we examined ten of the meiotic lines at intermediate stages of the meiotic bottleneck ( T-10 , 20 , 30 , 40 ) . All of the lines displayed spore viability similar to the parental line . In contrast , three of twenty vegetative lines displayed spore viabilities consistent with the accumulation of a single recessive lethal mutation . Such a result is consistent with vegetative lines accumulating heterozygous mutations ( see below ) . Vegetative and meiotic lines were examined for the presence of mutations using deep sequencing , PFGE , and CGH . To provide an estimate of vegetative and meiotic mutation rates in diploid yeast , whole genome paired end sequencing was performed for the mitotic 3B-87 and 4B-87 lines , and for the meiotic 3T-50 and 4T-50 lines ( Materials and Methods ) . For the 3B-87 , 4B-87 and 4T-50 lines a single haploid spore clone was isolated from a complete tetrad from the final bottlenecks , germinated and grown in culture . For the 3T-50 line three spores from a complete tetrad were germinated and grown in culture . The parental strain , EAY2531 , was sequenced as a diploid because no heterozygosities apart from the MAT locus were expected; none were detected by sequencing . We also sequenced the diploid genome of the 2B line at generation 52 ( 2B-52 , ∼1040 generations ) using a single end approach . The sequencing coverage is presented in Table S1 . For the vegetative lines , eight , six , and five base substitutions were identified in 3B-87 , 4B-87 , and 2B-52 , respectively ( Table 1 ) . The nineteen base substitutions were verified by Sanger sequencing of DNA isolated from 3B-87 , 4B-87 , and 2B-52 diploids ( Materials and Methods ) . This analysis also confirmed that sporulating the lines at the end of the bottlenecks did not introduce new mutations . All nineteen substitutions were heterozygous in the diploid lines; this was expected because they were propagated clonally in the absence of a sexual cycle . For the 3B-87 and 4B-87 lines half of the genome was sequenced because only one spore clone was analyzed; thus to determine the genome-wide mutation rate for these two lines , we multiplied by two the number of base substitutions detected . After this correction we estimate that the single base substitution rates in the vegetative 3B87 , 4B-87 and 2B-52 lines were 3 . 8×10−10 , 2 . 8×10−10 and 2 . 0×10−10 substitutions per base per cell division , respectively ( 24 , 483 , 546 bp genome at 96% coverage for 1740 ( 87 bottlenecks ) or 1040 ( 52 bottlenecks ) generations ) . The average of these rates , 2 . 9×10−10 per base per cell division , is very similar to values obtained by Lynch et al . [17] in a haploid mutation accumulation study ( 3 . 3×10−10 ) , and by Drake [36] who estimated base substitution mutation rates in haploid yeast at the CAN1 ( 1 . 7×10−10 ) and URA3 ( 2 . 8×10−10 ) loci . For the meiotic line 3T-50 the same five base substitutions were detected in genomic DNA isolated from each of the three sequenced spore clones . This information , in conjunction with Sanger sequencing from 3T-50 diploid cells , indicated that the five base substitutions were homozygous in the final bottleneck . The one spore sequenced from the 4T-50 line also contained five base substitutions ( Table 1 ) . Sanger sequencing from 4T-50 diploid cells indicated that these five base substitutions were also homozygous in the final bottleneck . Because all mutations were homozygous in the meiotic lines , we did not need to correct for the total number of base substitutions , even for the 4T-50 line where we only sequenced one spore . However , to determine the base substitution rate , we multiplied the number of base substitutions in each line by two to account for the loss of half of the base substitutions accumulated in the vegetative phase of the bottleneck during intra-tetrad mating ( see below ) . Based on this assumption , both lines showed the same base substitution rate , 3 . 9×10−10 per base per cell division ( 10 base substitutions per line in a 24 , 483 , 546 bp genome grown for 1 , 000 vegetative and 50 meiotic generations ) . This value is nearly identical to that obtained for the vegetative base substitution rate estimate . Most mutations in the vegetative and meiotic lines ( 17/29 ) were in coding regions and resulted in non-synonymous substitutions ( Table 1 ) . Of the nineteen base substitution mutations detected in vegetative lines , eight were transitions and eleven were transversions ( Table 1 ) . Twelve of these mutations resulted in a change from a G-C to an A-T base pair , whereas only five were in the opposite direction . For the ten base substitutions seen in the meiotic lines , four were transitions and six were transversions ( Table 1 ) . Seven of these resulted in a change from a G-C to an A-T base pair , whereas only two were in the opposite direction . The overall bias towards A-T base pairs was seen and discussed previously ( e . g . [17] , [37] , [38] ) . The fact that we did not observe significant differences between the base substitution rates of the mitotic and meiotic lines could reflect the relatively low number of meiotic ( 50 ) compared to vegetative divisions ( 1 , 000 ) in the meiotic bottlenecks . To estimate the upper limit of the meiotic mutation rate we simulated the occurrence of mutations given different meiotic mutation rates and taking into account the experimental setup . The rates obtained by simulation were compared to the observed rates to establish a range of meiotic mutation rates consistent with the observed values . We considered two scenarios in this analysis ( Figure S1 ) . In the first , mutations occurred prior to meiotic DNA replication and are thus present in two of the four chromatids of a homolog . In the second , mutations occur during or after meiotic DNA replication ( during double strand break repair ) and are present in only one of the four chromatids . In both scenarios we accounted for the spore self-mating frequency that was experimentally determined ( 17% , see below ) . As shown in Figure 2A and 2B and Figure S2A , in the first scenario the distribution of simulated mutations became statistically different from the observed meiotic rate ( P<0 . 05 ) when the simulated meiotic mutation rate ( μ ) was 30-fold higher than the vegetative rate ( m ) . This shows that the meiotic mutation rate is only consistent with the observed rates if it is within the range of being equal to zero to being 30-fold higher than the vegetative rate . In the second scenario ( Figure 2C and 2D , Figure S2B ) , the meiotic mutation rate can be around 55-fold higher than the vegetative rate and still be consistent with our observations; if it was higher than that we would have observed a difference between the rates of the two mutation accumulation schemes . Although our experiments do not allow exact determination of the meiotic mutation rate they show that this rate can be no higher than ∼1 . 74×10−8 per base per cell division in S . cerevisiae . To identify 1–2 nt in-del mutations , we aligned the sequencing reads obtained for all of the sequenced lines against the reference genome SK1 using the Novoalign software ( Materials and Methods; http://www . novocraft . com ) . Statistical methods were performed to identify high confidence 1–2 nt in-del mutations ( XM and CB , unpublished; Materials and Methods ) . We did not detect such in-dels in any of the sequenced lines . A second approach to identify in-dels by aligning the reads to the S288c sequenced genome also did not reveal any in-dels specific to the mutation accumulation lines ( see Materials and Methods ) . To search for intermediate-sized structural variants ( SV; >500 bp ) , we analyzed positional discrepancies between paired-end reads [39] and performed read depth coverage analysis [40] , [41] . The SV predictions were validated using real-time quantitative PCR ( qPCR ) , Southern blotting , and PCR ( Table S3; Figure S3; Materials and Methods ) . In paired end mapping , SVs larger than a cutoff of approximately 500–1 , 000 bp ( depending on the insert size distribution , see Materials and Methods ) can be identified . However , pair-end mapping did not identify SVs that were specific to the sequenced mutation accumulation lines . Read depth analysis can identify SVs larger than 900 bp ( see Materials and Methods ) . Only one of 55 potential SVs identified by read depth analysis was verified by both qPCR and Southern analysis ( Figure S3; data not shown ) . A region ( ∼3 . 0 KB ) that showed high similarity to the Ty3 element , a relatively rare class of retrotransposon present in yeast ( two copies in S288c; [42] ) , was present at higher abundance in 3T50 than in the parental strain , suggesting the gain of at least one copy . Southern analysis showed that a new Ty3 element was inserted into the ribosomal DNA cluster on chromosome XII in the 3T-50 isolate ( data not shown ) . The location of the retrotransposition was determined by PCR and Sanger sequencing ( Figure S3 ) . While we were successful in identifying a Ty3 retrotransposition event , it is important to note that our read depth analysis does not have the sensitivity to detect copy number variation associated with transposition of more abundant repetitive elements such as Ty1 or Ty2 ( ∼50 copies in S288c; [42] ) . It is also not possible to detect SVs of between three and 500 bp with our short-read data . However , the low number of intermediate sized SVs found is surprising given previous measures of gene duplication and gene loss in haploid mutation accumulation lines of yeast ( [17]; see Discussion ) . In addition to whole genome re-sequencing of specific mutation accumulation lines , we investigated the occurrence of gross chromosomal rearrangements in all vegetative ( 20 ) and meiotic ( 19 ) lines by using PFGE to resolve full-length chromosomes ( Figure S4 , Figure S5 ) . As summarized in Table 2 , the chromosomal rearrangements detected in the two strain sets were similar in both their high abundance ( ∼75% of lines had at least one visible size change ) and their large-scale deviation from the respective parental chromosomes ( ±10 to 40 KB ) . In both sub-culturing regimens , Chromosome IX was the least stable chromosome ( ∼50% of all size changes ) , with ten cases detected in the meiotic lines and eleven in the vegetative lines . While we frequently observed heterozygous changes in the vegetative lines ( i . e . two homologs of different size could be distinguished ) , in the meiotic lines , all but 1 of the 24 instances of the size changes were present in both homologs of the affected chromosome , presumably due to loss of heterozygosity through meiotic inbreeding ( see below ) . We also saw an increase in chromosome size in the meiotic lines ( seventeen chromosome sizes increased and seven decreased ) compared to the vegetative lines ( seven increased and ten decreased ) , but this difference was not statistically significant ( P = 0 . 11 , Fisher's Exact Test ) . Finally , we also noted that changes in the meiotic lines involved a more diverse set of chromosomes than in the vegetative lines ( seven chromosomes vs . three chromosomes , respectively ) . We used comparative genomic hybridization microarrays ( array CGH; [43] , [44] ) to investigate the molecular nature of the chromosomal rearrangements . This analysis revealed that the original diploid gene copy complement was maintained for nearly the entire genome in the seven mutation accumulation lines assayed , including all four sequenced lines ( ∼4 KB resolution; data not shown ) . The only exceptions were cases of copy number variation detected at Y′ subtelomeric regions . Consequently , we used high resolution PFGE ( Figure 3A ) to better visualize the chromosomal rearrangements in these lines , and conducted Southern analysis using the Y′ sequence as probe ( Figure 3B ) . This blot revealed that increases or decreases in chromosome size were always associated with a corresponding higher or lower intensity of the Y′ hybridization signal . This was clearly illustrated by chromosome I in the 5T-50 meiotic line , which is about 40 KB longer than the parental chromosome I , and showed a much stronger Y′ hybridization signal . Also consistent was the observation that the Y′ hybridization signal for chromosome IX in the parental strain was stronger compared to other chromosomes , suggesting the presence of an expanded multi-copy Y′ allele on chromosome IX . This last result suggests a mechanism for the high instability observed on this chromosome through unequal crossing over . We further investigated the involvement of Y′ sequences in the observed chromosome size variation by digesting full length chromosomal DNA with the MluI restriction endonuclease , which does not have recognition sites in Y′ , and therefore releases terminal chromosomal fragments . The MluI digested DNA was separated by size with PFGE and probed with Y′ to visualize the terminal fragments ( Figure 3C ) . This analysis uncovered additional cases of size variation that were too small in range to be resolved in chromosomal PFGE , and also narrowed down their occurrence to the regions near the ends of chromosomes . All seven strains analyzed displayed at least two chromosome ends of variant size . Taken together , our data strongly suggest that most of the chromosomal rearrangements that accumulated in the mutation accumulation lines were due to Y′ recombination . Since the rearranged regions did not span essential genes , this result also explains why spore viability remained high in the mutation accumulation lines despite the presence of chromosomal rearrangements . While we did not investigate the specific break point structure of the Y′ rearrangements , our data suggest that none of the rearrangements involved breakpoints at internal locations . First , all chromosome size changes were associated with a corresponding increase or decrease in the hybridization signal for the Y′ probe in PFGE/Southern analysis . Second , non-reciprocal translocations associated with copy number variation were not observed in the array CGH assay . Third , the high spore viability seen for the vast majority of lines ( except for those containing lethal heterozygous mutations ) suggest that reciprocal translocations did not occur; such events would have likely conferred reduced spore viability . Fourth , any reciprocal translocations that formed would have to be very close in size ( within 5 to 10 KB ) to the parental chromosomes . Lastly , paired-end analysis would have identified such breakpoints; none were identified . In addition to structural chromosomal aberrations , we also looked for changes in chromosome number using image tracing analysis of the PFGE profiles ( data not shown ) . This analysis showed that for the entire data set all chromosomal bands of unchanged size were present at the same intensity relative to the parental strain ( data not shown ) , indicating that aneuploidy never accumulated in any of the lines . The presence of homozygous base substitutions and structural variants in the meiotic lines can be explained by the initial appearance of heterozygous mutations that are fixed to homozygous in subsequent meiotic bottlenecks by inbreeding . Self-mating through HO-induced mating-type switching [45] will immediately lead to fixation or purging of a mutation while inter-spore mating would lead to fixation or purging only in a fraction of the possible mating combinations ( see below ) . To estimate the frequencies of self-mating and inter-spore mating , we inserted the kanMX and natMX markers at chromosome III at the ARS314 locus that is tightly linked ( 1 . 5 KB proximal ) to MAT in the diploid homothallic parent strain EAY2531 ( Figure 4A ) . The introduction of these drug markers is unlikely to affect the efficiency of MAT locus switching because the insertions are distal to the HO-induced DSB site . Consistent with this , single spores from strains containing the kanMX or natMX insertions near MAT were able to switch mating type and form diploids at frequencies similar to those from strains unmarked near the MAT locus ( data not shown ) . A diploid that forms by inter-spore mating will be resistant to both G418 and nourseothricin . A diploid formed by self-mating will be resistant to G418 or nourseothricin but not to both . Our analysis accounts for rare single crossovers ( double crossovers would not affect genotyping of the diploids ) that can occur between the drug markers and the MAT locus , yielding progeny resistant to only one drug but arising from inter-spore mating ( Table S4 ) . This was determined by creating haploid strains EAY2694 and EAY2697 in which drug markers were linked to MAT and the HO gene was disrupted . The genetic map distance between the drug markers and the MAT locus ( 1 . 5 KB physical distance ) was 1 . 0 cM , suggesting that the drug marker insertions would not have a major effect on the analysis ( Table S4 ) . Two independent diploid colonies were isolated from the single cell streak of each germinated tetrad colony of EAY2771 ( relevant genotype ARS314::kanMX/ARS314::natMX ) and tested for drug resistance to G418 and nourseothricin . Two different methods , streaking and microdissection , were performed with similar results; we obtained an inter-spore mating frequency of 82% and self-mating frequency of 18% ( Table 3 ) . Taking into account the crossover frequency between the drug-resistant markers and the MAT locus ( Table S4 ) , the revised estimates for inter-spore mating and self-mating were 83% and 17% respectively ( Figure 4B ) . Analysis of the intra-tetrad mating pattern also showed the presence of multiple types of mating within a single tetrad . For 18% of the tetrads analyzed , two single colonies arising from the same tetrad showed different patterns of drug resistance ( Table 3 ) . This indicates that the occurrence of one type of mating event does not prevent additional and different types of mating events within a single tetrad . The low frequency of self-mating indicates that it plays only a minor role in fixing mutations in our meiotic lines . The excess of homozygous mutations in the meiotic lines is likely due to random inter-spore mating during the meiotic bottlenecks . These analyses also suggested that the population size of the bottleneck in the meiotic lines is variable , between one and four . In our bottleneck scheme , if the formation of a diploid cell from a germinating tetrad occurs only by inter-spore mating , a heterozygous mutation unlinked to MAT has a 2/3 chance to remain heterozygous in the resulting diploid , and a 1/6 chance to become mutant homozygous or wild-type homozygous ( Figure S1A and S1B ) . After multiple rounds of meiosis followed by mating , half of the mutations that are initially heterozygous will become homozygous and half will be lost . Since we determined the proportion of inter-spore mating to be 83% , the probability of a mutation being fixed to homozygosity or lost after n rounds of meiosis by inter-spore mating is 1− ( 2/3×0 . 83 ) n . The diploids that formed by self-mating ( 17% ) will become either fixed or lost in a single round . Based on these calculations , a new heterozygous mutation has a probability of over 99% of being fixed or lost after nine meiotic bottlenecks . The above prediction was confirmed by sequencing five of the base substitutions identified in one of the meiotic lines ( 3T50 ) at intermediate stages ( 10 , 20 , 30 and 40 rounds ) of the bottleneck . All five base substitutions became homozygous mutant in 10 or fewer meiotic bottlenecks ( Table S5 ) . For two of the five mutations a heterozygous mutation could be seen in an intermediate bottleneck . Consistent with these results , in the simulation of the meiotic bottlenecks ( Figure 2 and Figure S2 ) the observed number of heterozygous SNPs is always considerably low relative to homozygous SNPs . Our data show that heterozygous mutations will persist when propagated vegetatively but will fix very rapidly when propagated in alternating cycles of vegetative and meiotic growth due to self and spore-spore mating within a tetrad . This information will be useful to estimate how often S . cerevisiae undergoes a meiotic cycle based on the level of heterozygosity in wild populations ( e . g . [34] , [46] ) .
The single base substitution rates in the meiotic lines and vegetative lines were nearly identical to each other and to rates obtained previously by Lynch et al . [17] working with haploid lines . Although mutagenic effects of meiosis were not observed in our meiotic lines as measured by spore viability , it is possible that at the level of a single division , meiosis is more error prone than mitosis . Because our lines alternated between 20 vegetative generations and one meiotic generation , our estimates for meiotic mutation are less precise . Based on simulations shown in Figure 2 , we can assign upper limits ( 25 to 55-fold ) on increased base mutation rates in meiosis compared to vegetative growth . A more precise genome-wide estimate would likely require a large-scale deep sequencing analysis that involves sequencing DNA from thousands of independent spore clones from a single round of meiosis . However , based on recent work in S . paradoxus suggesting that a meiotic cycle occurs only once per 1 , 000 vegetative cycles [34] , it's not clear if a meiotic mutation rate at the upper limit as predicted by previous studies [10]–[12] would significantly impact fitness in baker's yeast . During meiosis approximately 140 to 170 double-strand breaks ( DSBs ) are induced in a single cell [48] . These breaks are repaired through mechanisms that involve roughly 0 . 8 to 1 . 9 KB of DNA synthesis [49] . Thus as much as 500 KB of DNA is re-synthesized during DSB repair in a single cell during meiosis . Strathern et al . [11] have estimated that the misincorporation rate of the DNA polymerase ( s ) associated with mitotic DSB repair is 10−6 to 10−5; such a high rate could be responsible for the high meiotic mutation rates observed by Magni and von Borstel [10] . In the meiotic bottleneck performed in this study ( 50 meioses , with only 25% of the mutations recovered because repair synthesis is thought to occur on only one of the four chromosomes , and half are lost due to mating ) , one would expect in each line between 3 ( 1×10−6 rate ) and 30 ( 1×10−5 rate ) mutations associated with meiosis . Our data are compatible with a DNA polymerase misincorporation rate of 1×10−6 , but suggest that previous upper-end estimates are too high . Alternatively , the polymerases associated with meiotic DSB repair are more accurate than those active in mitotic DSB repair , or DNA mismatch repair could more often excise DNA synthesis errors in meiotic DSB repair than in mitotic DSB repair . No SVs larger than 500 bp ( we could not detect SVs between 3 and 500 bp ) were detected in single-copy coding regions or other single-copy sequences despite using methods ( read depth coverage , PFGE , array CGH ) that are highly sensitive to a large range of structural variant sizes . These results are in contrast to the findings of Lynch et al . [17] who observed that the majority of the mutational changes in the haploid vegetative lines were structural variations involving copy number changes mediated by non-allelic homologous recombination ( NAHR ) between dispersed repeat elements distributed throughout the genome . Another important difference is that our diploid lines did not show the genomic instabilities that were frequently observed by Lynch et al . [17] at different stages in their bottleneck analysis . In fact , the haploid lines in Lynch et al . [17] rapidly accumulated whole chromosome gains to become in effect diploids within a few hundred generations , at which point the relative chromosome number stabilized . A reverse trend was observed when yeast tetraploids submitted to a subculturing regimen progressively lost chromosomes until a stable diploid state was reached [50] . These results point to a remarkable stability of the yeast diploid genome and to the possibility that diploid genomes are inherently more resistant to chromosomal rearrangements than haploids . This occurs despite the fact that diploids are able to tolerate the deletions of large regions spanning essential genes that would otherwise be lethal in a haploid . The availability of a homologous chromosome template at all times during the cell cycle is likely to improve the fidelity of repair of chromosomal breaks , since most mitotic crossover events in diploid yeast can be explained by precursor double-strand break lesions that occur during G1 , before genome replication [51] . A haploid genome would be ill equipped to repair such breaks , possibly leading to SVs similar to those observed by Lynch et al [17] . Small in-del mutations are thought to occur at high rates in homopolymeric runs due to replication slippage [52] , [53] . As described above , we did not identify 1–2 nt in-del mutations in any of the sequenced lines . One possibility is that our methods were not sensitive enough to detect in-dels in homopolymeric tracts . Such mutations can be identified only if the entire tract and unique sequences flanking both sides are present in the short read sequence , and the alignment program can map the in-del . However , in another study our methods have been successful in identifying in-dels in five to thirteen bp homopolymeric tracts located in single copy genes ( XM and CB , unpublished data ) . Previous estimates for frameshift mutation rates in homopolymeric tracts at the LYS2 locus in yeast were 3 . 3×10−9 , 16×10−9 , and 47×10−9 for A5 , A8 , and A10 runs , respectively [52] . In a comparison of N10 tracts at a single locus , Gragg et al . [53] observed rates that varied from 24×10−9 for A10 tracts to 10 , 500×10−9 for G10 tracts . If we assume that all in-dels occur in homopolymeric tracts in the diploid yeast genome ( 154 , 850 , 98% of which are ≤10 nt ) , then the rate of in-dels in our generation 87 bottleneck lines is <3 . 7×10−9 , which appears lower than previous mutation rate estimates for homopolymeric tracts of five to ten nucleotides [17] , [52] . While the single copy regions of the genome were highly stable , our subculturing lines showed widespread structural variation in the regions near chromosome ends with low gene content , namely in the Y′ subtelomeric repeats . Such repeats are highly variable between yeast strains and have been shown to recombine ectopically both in vegetative and meiotic cells ( reviewed in [54] ) . Analogous dynamic structures have been well characterized in human subtelomeres as well [55] . The high rate of subtelomeric recombination mediated by both homologous and non-homologous mechanisms is thought to be responsible for the remarkable diversity of subtelomeric configurations that exist between individuals , many of which have been implicated in disease processes [56] . Taken together , our results are consistent with the high rate of structural variation at subtelomeres , and support the proposal that the peripheral regions of chromosomes are much more plastic than the gene-rich central sections of the diploid genome where structural rearrangements are associated with more severe phenotypic consequences [57] , [58] . It is important to note that there are methodological differences between our study and that of Lynch et al . [17] in the sequencing technologies used . Their analysis involved longer sequencing reads but at lower coverage ( ∼5-fold , ∼50% of the genome ) ; we obtained much deeper coverage ( 40 to 60-fold , ∼95% of the genome ) . Our sequencing approach prevented us from accurately detecting structural variations of between 3 and 500 bp . However , since the differences seen in the two studies primarily involved structural variations greater than 1 KB , the use of different technologies should not be a factor in interpreting the two data sets . Intra-tetrad mating appears to be a major component of the sexual life cycle of most yeasts , while the frequency of outcrossing in S . cerevisiae is estimated to be very low , once every 50 , 000 divisions [46] , [59] . Many natural isolates of yeast are homothallic ( HO ) and are capable of switching mating type ( [45]; reviewed in [60] ) . We showed by linkage analysis that 83% of intra-tetrad matings in homothallic S . cerevisiae tetrads occur by inter-spore mating . This is consistent with the high frequency of inter-spore mating ( 94% ) that was previously inferred in S . paradoxus tetrads by population genetics based approaches [34] . The high frequency of spore-spore mating seen in S . cerevisiae may be due to the presence of inter-spore bridges that are maintained within a tetrad [61] . Self-mating might also be impeded by the requirement that the mother cell undergo two divisions before it can switch mating type [45] . Intra-tetrad mating is expected to create a drive towards homozygosity of mutations . The absence of empirical estimates for the relative frequencies of different modes of mating in a tetrad , and the experimental difficulty of tracing mutations on a genome wide scale following mating have led to considerable theory [31] , [62] , but little direct evidence for the relationship between mutation heterozygosity and mating pattern exists . Given the advantages for the maintenance of heterozygosity in populations [63] , one might expect inter-spore mating to have a selective advantage over self-mating , since heterozygosity is lost only by a third during inter-spore mating whereas it is completely lost by self-mating [30] . By tracking mutations identified through whole genome sequencing of the meiotic mutation accumulation lines and using experimentally determined estimates of mating patterns within a tetrad , we showed that most new mutations , including base substitutions and structural variations , can go to fixation very rapidly , in less than ten rounds of meiosis and inbreeding . The ratio between mitotic and meiotic cycles in wild populations of S . cerevisiae is not known , although in S . paradoxus , population genetics approaches have determined that it undergoes a sexual cycle approximately once every 1 , 000 asexual cycles [34] . The information presented in this study should encourage the use of population genetic approaches to estimate how often S . cerevisiae undergoes a meiotic cycle ( e . g . [34] , [46] ) . Our data provided an estimated base substitution rate of 2 . 9×10−10 ( per base per cell division ) for vegetative growth in diploid baker's yeast . This analysis also showed that the meiotic mutation rate in baker's yeast is within the range of being equal to zero to being 55 times higher than the vegetative rate . We observed a large number of structural variations at subtelomeric regions in vegetative and meiotic lines and did not appear to affect spore viability . Only one structural variant was observed at a non-telomeric location , and no changes in ploidy were seen . Together , these data illustrate the remarkable stability of the baker's yeast diploid genome in the vegetative and meiotic cell cycles .
Yeast strains were grown on yeast extract-peptone-dextrose ( YPD ) medium [64] . When required , YPD medium was supplemented with Geneticin ( G418 , Invitrogen , San Diego ) and nourseothricin ( Werner BioAgents , Germany ) as described previously [65] , [66] . Sporulation medium was prepared as described in Argueso et al . [67] . Mutation accumulation lines were initiated with the SK1 strain EAY2531 ( MATa/MATalpha , HO/HO , ura3Δ::hisG/ura3Δ::hisG , leu2::hisG/leu2::hisG , lys2/lys2; [35] ) . We made this parental strain by isolating a diploid strain from a single homothallic spore derived from NKY730 ( kindly provided by Nancy Kleckner , same genotype as EAY2531 ) . Like NKY730 , EAY2531 sporulates rapidly and with high efficiency and spore viability ( 96% ) . EAY2531 was streaked to single cells on solid YPD media and after 48 hrs of growth at 30°C , 20 single colonies were split into two sets of lines . One set of 20 lines was designated the vegetative bottleneck line “B” and the other , “T” was designated to undergo vegetative and meiotic cycles as explained below . All strains were grown to saturation in 100 ml of YPD medium at 30°C and high quality DNA was extracted using a QIAGEN Genomic Tip according to the manufacturer's instructions . 5 µg of genomic DNA were fragmented using a Covaris DNA shearer and size-selected to ∼300 bp in a 2% agarose gel . Sequencing libraries were generated using an Illumina Genomic DNA Sample Prep Kit , according to the manufacturer's protocol . To increase coverage and allow detection of in-dels and SVs , all strains were sequenced paired-end with 36 nt reads using an Illumina Genome Analyzer GAII . Sequencing information for each strains is shown in Table S1 . The sequences data were submitted to the European Read Archive ( accession number ERA007227 ) . Short 36 nt reads of the parental EAY2531 strain , one lane of paired-end ( 8 , 101 , 474 pairs ) and two lanes of single-end ( 8 , 295 , 633 reads ) sequence , were used for de-novo assembly using Velvet [69] and ABySS [70] separately . First , the optimal k-mer size , for both tools , was determined by scanning the whole parameter space ( kmer size from 11 nt to 31 nt ) for the best assembly . The N50 , the size of the longest contig , the overall number of nucleotides in the assembly , and the number of generated contigs were used as metrics to identify the best assembly . The final assembly was done with a kmer size of 27 and 29 for Velvet and ABySS , respectively . The generated contigs were then combined using the minimus2 software [71] . In total , 1 , 139 contigs were assembled with an N50 of 36 , 291 bp . These contigs were then aligned to the SK1 genome sequence of the Saccharomyces Genome Resequencing Project ( SGRP; [72] ) using BLAST [73] . Gaps in the SGRP SK1 genome were filled with the corresponding sequences from the parental EAY2531 strain , and SNP and small in-del errors were corrected . In total , 56 gaps were filled . The new SK1 haploid genome sequence has 12 , 241 , 773 bp , covering 96% of the whole S288c nuclear genome . This SK1 sequence was used as a reference for further analyses . The reference sequence can be downloaded and also searched using BLAST at http://steinmetzlab . embl . de/SK1 . Short reads from the parental strain and the mutation accumulation lines were separately mapped to the SK1 references genome using the MAQ software [74] . Two mismatches were allowed for short-read alignment . For each strain , base substitutions and short in-dels were detected using the default filtering parameters of MAQ [74] . Detected polymorphisms in the mutation accumulation lines were compared to those detected in the parental strain to define strain-specific mutations . These mutations were manually checked in the alignment and finally confirmed by Sanger sequencing the 3B-87 , 4B-87 , 3T-50 , and 4T-50 diploid lines . For the 2B-52 line , two of the five mutations were confirmed by sequencing three haploid spores derived from the 2B52 line . The three others were confirmed by sequencing the 2B52 diploid . We also used the Novoalign ( v2 . 05 . 16; http://www . novocraft . com/ ) software to identify in-dels in sequenced lines using the reference SK1 genome . Since none were observed , we developed a second approach to identify in-dels . We aligned the reads directly to the S288c sequenced genome ( http://www . yeastgenome . org ) , which is 0 . 7% sequence divergent from the SK1 genome . We detected approximately 9000 in-dels; however most of these were seen in all of the lines , indicating that they were likely due to sequence differences between the S288c and SK1 genomes . After discarding in-dels that were detected in 9 or all 10 sequencing runs of the parental and bottleneck lines , approximately 1 , 000 in-dels remained . All of the short reads that covered these in-del sites were aligned back to the SK1 assembled genome . None of these in-del calls could be confirmed after alignment , indicating that they resulted from sequence differences between the S288c and SK1 genomes . To estimate the upper limit of the meiotic mutation rate , the occurrence of base substitutions was simulated in silico taking into account the experimental setup . For each of the 20 mitoses that occurred before each meiotic bottleneck in the “T” lines , a random number of base substitutions was generated given the observed mitotic mutation rate ( 2 . 9×10−10 ) and the size of the diploid nuclear genome ( 24 , 483 , 546 bp ) . Then , base substitutions in one meiosis were generated given different putative meiotic mutation rates ( μ ) . Two scenarios for the occurrence of meiotic mutations were considered: one in which mutations occur before meiotic replication and are therefore present in both strands of one of the two sister chromatids of a chromosome , and one in which mutations occur during or after meiotic replication and are present in only one of the strands of a sister chromatid of a chromosome . Once the mutations for 20 mitoses and one meiosis have been generated as described above , the meiotic bottleneck is simulated . The spores that will undergo spore-spore mating or self-mating to form the diploid cell were chosen randomly from the four chromatids considering the observed frequencies of intra-spore mating ( 83% ) and self-mating ( 17% ) . Since one or two spores were chosen in a single meiotic bottleneck , heterozygous base substitutions can be fixed to homozygous , persist as heterozygous or be lost for the next set of mitoses and meiotic bottleneck ( Figure S1 ) . The probability of two base substitutions occurring at the same position is very low , therefore the total number of base substitutions observed at the end of the 20 mitoses and one meiosis in the spores that will form the ongoing diploid cell equals the sum of base substitutions generated in each cell division in such spores . The set of 20 mitotic divisions plus one meiosis were then repeated 50 times to simulate the 50 meiotic bottlenecks that a single line underwent . For each of the tested meiotic mutation rates ( μ ) , 10 , 000 of the processes consisting of 50 bottlenecks were simulated and the distribution of the resulting number of base substitutions was recorded . The P value of the difference between the simulated distribution of base substitutions and the observed rate in the meiotic line was estimated as the frequency of simulations with equal or lower number of base substitutions than the average of the observed values . Paired end mapping was carried out using the PEMer algorithm with default parameters [39] . For the read-depth analysis , paired-end sequencing reads were aligned to the SK1 reference genome assembly using Novoalign ( v2 . 05 . 16; http://www . novocraft . com/; parameters used: -rRandom -Q 0 -R 5 ) . Only reads with an alignment quality of >125 were used for downstream analysis ( >9 . 6 million high-quality reads for each sample ) . The number of aligned reads was then counted in consecutive genomic windows of predefined size . Windows of between 100 and 400 bp were tested and a final size of 200 bp ( 100 bp overlap ) was selected since it achieved a good trade-off between resolution and noise-level . Read-depth signals were scaled using quantile normalization [75] . For each window the log2 of the ratio between the read-depth of each of the mutation accumulation lines and the read-depth of the parent was calculated . To reduce the noise level sample specific GC-correction was performed and windows with less reads than the median ( read_depth ) – 2×standard deviation ( read_depth ) were discarded . To account for the remaining waviness of the data , local regression ( LOESS ) was performed [76] with a span representing a region of 20 KB . To identify consecutive windows that show abnormal log2-ratios we used two approaches , CNV-seq [40] and DEseq ( http://www . bioconductor . org/packages/2 . 6/bioc/html/DESeq . html ) . CNV-seq was used with a log2-ratio threshold of ±0 . 48 and a P-value threshold of 1×10−37; only regions larger that 900 bp were considered . DEseq was employed without log2-ratio threshold and a P-value cutoff of 0 . 0001 . Furthermore , when using DEseq , at least two abnormal windows per SV were required at most 1 KB apart from each other . 55 putative SVs in 52 different loci that ranked highest in the read-depth analysis were further analyzed by qPCR ( Table S3 ) . qPCR was performed in an ABI 7500 thermocycler using SYBR Green and standard settings ( Applied Biosystems ) . Reactions were performed at least in triplicates and the parental and target samples were always ran in the same plate for a given primer pair . Among-sample variation in the amount of DNA used in each reaction was normalized using independent primers for the single-copy genes BUD23 and ERG1 . The relative copy number difference between the mutation accumulation line and the parent was calculated as the CT difference between both lines minus the CT difference in the control regions BUD23 or ERG1 . PFGE was conducted using a BioRad Contour-clamped homogeneous electric field ( CHEF ) Mapper XA system . Agarose-embedded chromosomal DNA preparation and running conditions were performed as described previously [77] . The genomic DNA used for array CGH was purified from agarose plugs prepared for PFGE , using a procedure modified from the QIAGEN QIAquick Gel Purification Kit . Briefly , four ∼70 µl agarose plugs per sample were dissolved in 840 µl of QIAGEN QG buffer . The DNA in this solution was fragmented through sonication to a size of 1–2 KB , and 280 µl of isopropanol were added . The mixture was bound to QIAquick columns , washed with QIAGEN PE buffer , and eluted in 32 µl of QIAGEN EB buffer . This procedure yielded 2–3 µg of fragmented DNA which was labeled and hybridized for array CGH assays as described previously [78] . EAY2771 ( HO/HO , ars314::kanMX MATa/ars314::natMX MATalpha , ura3Δ::hisG/ura3Δ::hisG , leu2::hisG/leu2::hisG , lys2/lys2 ) was constructed by sequentially inserting kanMX and natMX drug resistance markers into identical positions at ARS314 in EAY2531 . This locus is 1 . 5 KB proximal to the MAT locus . 184 tetrads obtained by sporulating EAY2771 were placed by microdissection at unique positions on a YPD plate . All 184 tetrads germinated and formed colonies on YPD . The intra-tetrad mating pattern was determined using two approaches . In the first approach , 100 of the colonies were re-streaked on YPD plates to single cells . Two of the resulting colonies were patched onto YPD-G418 and YPD-nourseothricin to assess antibiotic resistance , and onto sporulation plates to assess ploidy ( all were diploids ) . In the second approach , the remaining 84 colonies were streaked onto YPD to single cells and two unbudded cells from each original colony were isolated under the dissection microscope . These single cells were incubated on YPD to form colonies . 145 of these cells formed colonies . The resulting colonies were patched onto YPD plates containing G418 or nourseothricin to assess antibiotic resistance , and onto sporulation plates to assess ploidy ( all were diploids ) . The second approach was performed to eliminate the possibility of closely spaced multiple cells from the restreak giving rise to single isolated colonies . EAY2775 ( ars314::kanMX MATa/ars314::natMX MATalpha , ho::hisG-URA3-hisG/HO , ura3Δ::hisG/ura3Δ::hisG , leu2::hisG/leu2::hisG , lys2/lys2 ) is a derivative of EAY2771 in which the HO gene was disrupted with the hisG-URA3-hisG marker . Haploid segregants of EAY2775 , EAY2694 ( ars314::kanMX MATa , ho::hisG-URA3-hisG ) and EAY2697 ( ars314::natMX MATalpha , ho::hisG-URA3-hisG ) , were mated on complete plates for four hours and then transferred to sporulation medium for 48 hrs . Tetrads were dissected on YPD medium and incubated at 30°C for 48 hrs . Spore clones were replica plated onto selective media and mating testers and segregation data from each replica was analyzed using the RANA software [67] . The kanMX and natMX markers ( EAY2771 ) each showed 2∶2 segregation and segregated independently from each other in all tetrads analyzed . No gene conversion events involving the drug resistance markers were seen . | Mutations result from errors that occur during DNA metabolism . They provide the raw materials for evolution , can affect organism fitness , and have been shown to accumulate in organisms during asexual growth . During a sexual life cycle , mutations can be removed by recombination and mating . While such removal is thought to provide a fitness advantage , studies have shown that recombination itself is mutagenic . To examine if the mutation rate in an organism differs during asexual and sexual cycles , we sequenced the entire nuclear genome of lines of diploid baker's yeast that underwent only asexual growth , or alternating cycles of asexual and sexual growth . The estimated rate of base substitutions in the vegetative lines was extremely low ( 2 . 9×10−10 base substitutions per base per cell generation ) and the meiotic mutation rate is within the range of being equal to zero to being 55 times higher than the vegetative rate . Interestingly , we observed a large number of changes in the ends of chromosomes in the asexual and sexual cycles that did not affect fitness; changes at other locations were very rare , suggesting a remarkable genome stability of diploid baker's yeast . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/dna",
"replication",
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"molecular",
"biology/recombination",
"molecular",
"biology/chromosome",
"structure",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/bioinformatics",
"molecular",
"biology/dna",
"repair",
"genetics",
"and",
"genomics/cancer",
"genetics"
] | 2010 | The Baker's Yeast Diploid Genome Is Remarkably Stable in Vegetative Growth and Meiosis |
MicroRNAs are universal post-transcriptional regulators in genomes . They have the ability of buffering gene expressional programs , contributing to robustness of biological systems and playing important roles in development , physiology and diseases . Here , we identified a microRNA , miR-125a , as a positive regulator of granulopoiesis . MiR125a knockout mice show reduced infiltration of neutrophils in the lung and alleviated tissue destruction after endotoxin challenge as a consequence of decreased neutrophil numbers . Furthermore , we demonstrated that this significant reduction of neutrophils was due to impaired development of granulocyte precursors to mature neutrophils in an intrinsic manner . We showed that Socs3 , a critical repressor for granulopoiesis , was a target of miR-125a . Overall , our study revealed a new microRNA regulating granulocyte development and supported a model in which miR-125a acted as a fine-tuner of granulopoiesis .
Neutrophils , also known as polymorphonuclear leukocytes ( PMNs ) , are the most abundant granulocytes which play a crucial role in immune defense and inflammatory reaction . Given that the post-mitotic nature of mature neutrophils , they have short lives about only a few days [1] and need to be regenerated constantly through granulopoiesis , a part of hematopoiesis occurring in the bone marrow of adult mammals . During granulopoiesis , hematopoietic stem cells , at the top of the hematopoietic hierarchy , produce multilineage progenitors and precursors-common myeloid progenitors ( CMP ) and subsequently granulocyte-monocyte progenitors ( GMP ) which differentiate into mature granulocytes including eosinophils , basophils and neutrophils [2] . In general , granulopoiesis is in a basal physiological condition . However , emergency granulopoiesis can be rapidly induced to produce large number of neutrophils if severe systemic infection occurs [3] . Hematopoiesis is regulated by a group of cytokines . G-CSF is one of the major cytokine that regulates cell proliferation , differentiation and survival during the neutrophil lineage commitment [4 , 5] . The receptor of G-CSF is mainly expressed in granulocytic progenitor cells and mature neutrophils [6] . The binding of G-CSF to its receptor triggers receptor dimerization and tyrosine phosphorylation of JAK1 , JAK2 and TYK2 , which belong to the Janus family of protein tyrosine kinases ( JAKs ) [7] . These then phosphorylate residues in the cytosolic part of the G-CSF receptor and subsequently activate mitogen-activated protein ( MAP ) kinase like ERK pathway [8] and the signal transducers and activators of transcription ( STATs ) including STAT1 and STAT3 ( 4 , 10 ) . SOCS3 , as the major repressor of G-CSF signaling , belongs to the suppressor of cytokine signaling ( SOCS ) family of proteins [9] , which can be recruited to phosphorylated cytokine receptors and inhibit JAK catalytic activity and subsequently inhibit activation of ERK and STATs . Moreover , mice with Socs3 conditionally knocked out in hematopoietic cells [10 , 11] develop neutrophilia and inflammatory pathologies . MicroRNAs ( miRs or miRNAs ) are universal post-transcriptional regulators in animals and plants . Primary miRNAs are first transcribed by RNA polymerase II or III and are then excised to mature miRNAs ( ~22 nucleotide ) that bind to 3’ untranslated regions ( UTR ) of their target mRNAs to silence gene expression [12] . More than 1000 miRNA genes have been identified in mammalian genomes [13] . And over 60% of protein-coding genes could be targeted by miRNAs according to computational prediction [14] . Due to their specific features , miRNAs have the ability of buffering gene expression programs and contributing to the robustness of biological systems [15] . Thus they play important regulatory roles in different biological processes . Decades of researches have shown that miRNAs involve in mammalian blood cell development and function [16] . For instance , miR-181a was found to modulate T cell selection [17] and miR-150 was identified as a controller of B cell development [18–20] as well as megakaryocytic versus erythrocytic lineage commitment [21] . In addition , miR-223 , which was found highly expressed in neutrophils , played a role in regulating the proliferation of granulocyte progenitors and also mediated the inflammatory function of neutrophils [22 , 23] . MiR-125a and miR-125b belong to the miR-125 family , which play a crucial role in many different cellular processes including cell differentiation , proliferation and apoptosis [24] . In order to systematically study the function of miR-125a in vivo , we developed miR-125a knockout mice . We examined the hematopoiesis of these mice and found fewer neutrophils in both bone marrow and peripheral blood in the absence of miR-125a . As a consequence of decreased number of neutrophils , MiR125a knockout mice were demonstrated with reduced infiltration of neutrophils in the lung and alleviated tissue destruction in an endotoxin challenge model . Furthermore , we found out that the reduction of neutrophils was due to impaired proliferation of immature granulocyte to mature neutrophils in an intrinsic manner . We showed that Socs3 , a critical repressor for granulopoiesis , was a target of miR-125a . Together , these results suggest that miR-125a is an important regulator of basal granulopoiesis .
To fully understand the physiological role of miR-125a in vivo , we generated the MiR125a knockout mice as previously described [25] . These mice are fertile , born at the expected mendelian ratio , and not shown any abnormalities during their growth . However , we found that the white blood cell differential count revealed decreased numbers of neutrophils in MiR125a-/- mice ( 1 . 4 ± 0 . 3 x 106cells/mL versus 2 . 2 ± 0 . 4 x 106 cells/mL ) ( p<0 . 0001 ) while other mature hematopoietic lineage cells including other granulocytes ( eosinophils and basophils ) were normal ( Table 1 ) . Flow cytometry analyses of neutrophils in the bone marrow and peripheral blood confirm these results ( Fig 1A ) . Next we did a bone marrow transfer assay to find out whether reduced granulopoiesis in MiR125a-/- mice are due to impaired cell-autonomous development or altered cytokine production from the bone marrow stromal cells . We found that decreased number of neutrophils reconstituted with MiR125a-/- bone marrow cells was both in MiR125a+/+ and MiR125a-/- recipients ( Fig 1B ) . These results demonstrate that miR-125a contributes to reduced granulopoiesis in a cell-autonomous way . In addition , morphological analysis shows that neutrophils in MiR125a-/- mice are as mature as those in wild-type mice ( Fig 1C ) . We then examined the expression of miR-125a in different stages of myeloid development and found that miR-125a was highly expressed in hematopoietic stem cells and decreased during maturation of myeloid progenitor cells , indicating that miR-125a may be involved in regulating granulocyte development ( Fig 1D ) . In order to examine whether miR-125a also plays a role in regulating neutrophil function , we tested the ability of activation , migration and killing pathogens between wild-type and MiR125a-/- neutrophils . Gene expression profiling data of bone marrow neutrophils stimulated with gram-negative bacterial lipopolysaccharide ( LPS ) showed that most of inflammatory factors and chemokines were induced equally either from MiR125a-/- or MiR125a+/+ mice ( S1A Fig ) . Then in vitro transwell assay showed MiR125a-/- neutrophils had no detectable abnormality in fMLP or CXCL1 or CXCL2-dependent chemotaxis and migration ( S1B Fig ) . We then used phorbol myristate acetate ( PMA ) or LPS to stimulate neutrophils and measured the production of reactive oxygen metabolites , which were important for neutrophils to kill pathogens . FACS analysis revealed no difference in the release of reactive oxygen species between wild-type and knock-out neutrophils ( S1C Fig ) . Furthermore in vitro killing assay also demonstrated MiR125a-/- neutrophils had normal ability to clear bacteria and fungi ( S1D Fig ) . Neutrophils are known to be recruited at inflammatory tissue sites and play a critical role in sepsis and tissue damage [26] . We therefore performed experimental endotoxaemia by injecting a sub-lethal intraperitoneal dose of LPS to MiR125a-/- mice for 24 hours and measured neutrophil infiltration in the lungs by flow cytometry . Lungs of MiR125a-/- mice accumulated fewer neutrophils than those of MiR125a+/+ mice ( Fig 2A ) . In addition , we checked the lung sections of MiR125a-/- and wild-type mice . Consistently with the FACS analysis , lungs of MiR125a-/- mice show less severe histopathological change , including congestion ( hyperplasia of alveolar walls and alveolar collapse ) , edema ( pulmonary interstitial edema ) , inflammation ( neutrophil infiltration ) and hemorrhage ( engorgement of the capillaries ) ( Fig 2B ) . We also found MiR125a-/- mice had significantly reduced serum amounts of aspartate aminotransferase ( ALT ) , blood urea nitrogen ( BUN ) , creatine kinase ( CK ) and creatinine ( CREA ) , which were indicators for organ damages ( Fig 2C ) . We next challenged both MiR125a-/- and wild-type mice with a lethal dose of LPS . We observed that MiR125a-/- mice were more resistant to lethal septic shock ( Fig 2D ) . However , serum concentrations of inflammatory cytokine IL-6 and TNF-α during sepsis were similar ( Fig 2E ) . In addition , normal Il6 and Tnfa mRNAs were expressed in peritoneal macrophages and bone marrow-derived macrophages after stimulation with LPS ( S2 Fig ) . To further study whether there is any macrophage involvement , we depleted endogenous macrophages by using clodronate liposomes in wild-type mice and transplanted with MiR125a+/+ or MiR125a-/- bone marrow-derived macrophages . Then we administrated these mice with the lethal dose of LPS . Results did not show any difference in mortality ( Fig 2F ) . These results implied that cytokine production induced by Toll-like receptors on macrophages did not contribute to resistance to LPS in MiR125a-/- mice . Thus resistance to a lethal dose of LPS and decreased neutrophils in the lungs with endotoxaemia in MiR125a-/- mice are likely caused by reduced granulopoiesis . To study the mechanism of decreased neutrophil numbers in MiR125a-/- mice , we performed flow cytometry analysis on bone marrow cells in both wild type ( WT ) and knockout ( KO ) mice to examine whether the frequency of progenitor cells was disturbed . We found that the numbers of myeloid progenitors did not change ( Fig 3A ) . We then performed colony forming assays on methylcellulose and analyzed them for myeloid precursors in complete medium . There is no significant difference in the frequency of myeloid precursors and numbers of granulocyte colonies ( Fig 3B ) . For greater precision , we performed colony assays in the medium only containing variant concentrations of G-CSF and found that there was also no change in colony numbers ( Fig 3C ) . However , we did notice that colonies from mutant mice were smaller and the cell number in one colony was less than those found in control mice ( S3 Fig ) . Thus we sorted Lin-Sca1-c-Kit+CD34hiCD16/32hi GMPs by FACS and estimated their developmental capacity in a CFU assay . We also found the colony number did not change ( Fig 3D ) but the colony size and the cell number per colony from MiR125a-deficient GMPs decreased in the presence of G-CSF ( Fig 3E–3G ) . Thus , it suggested that the development of granulocyte progenitors might be impaired in MiR125a-/- mice . Since the number of granulocyte progenitors remained unchanged , it would appear that reduction of neutrophils only was due to increased cell death or impaired proliferation from granulocyte progenitors to mature neutrophils . To test the first possibility , we examined cell death rate of Ly6Ghi cells from bone marrow by staining them with Annexin V and propidium iodide . We found no difference in the rate of cell death between MiR125a-/- and wild-type mice ( Fig 4A ) . We then performed in vivo BrdU-pulsing assays to analyze neutrophils generation in bone marrow ( Fig 4B ) and spleen ( Fig 4C ) . Flow cytometry results showed that neutrophils from MiR125a-/- mice incorporated less BrdU than wild-type mice , indicating that cell proliferation had decreased during the differentiation of granulocyte progenitors into neutrophils . It has been reported that CD11b+ Gr-1+ neutrophils in bone marrow are composed of three populations , including CD11bhi Gr-1hi cells ( mature Neu ) , CD11blowGr-1hi cells ( immature Neu ) and CD11bintGr-1int cells ( promyelocytes/myelocytes ) [27–29] . According to this , we found the percentage of immature neutrophils was significantly lower in the bone marrow of MiR125a-deficient mice while the percentages of promyelocytes/myelocytes and mature neutrophils had no change ( Fig 5A ) . In addition , we found BrdU-incorporating cells in the population of immature and mature neutrophils were significantly lower in MiR125a KO mice compared with WT controls while the population of promyelocytes/myelocytes had no change ( Fig 5B ) . Because of post-mitotic nature of mature neutrophils , these BrdU-incorporating mature neutrophils mostly came from BrdU-incorporating immature neutrophils during their last division . Thus we deduced that the neutropenia of MiR125a-deficient mice could be due to reduced cell proliferation of CD11blowGr-1hi immature neutrophils . As G-CSF is the major cytokine during granulocyte differentiation , we purified neutrophils from bone marrow cells and stimulated them with variant concentrations of G-CSF and counted the cell number after 24 hours . We found that the survival number of bone marrow neutrophils from wild-type mice increased substantially with increased G-CSF concentration while bone marrow neutrophils from MiR125a-/- mice did not increase in number ( Fig 6A ) . We then analyzed apoptosis percentage and BrdU-incorporated cell ratios in response to G-CSF . In accordance with the observation in vivo , the amount of BrdU-incorporation was less in the absence of miR-125a ( Fig 6B ) while the apoptosis percentage has no change ( Fig 6C ) . In addition , we found the mRNA levels of Gcsfr and several essential trancriptional factors for granulopoiesis like Pu . 1 , Gata-1 , Cebpa , Cebpb and Cebpe did not change ( S4 Fig ) . These results suggest that decreased cell proliferation in MiR125a-deficient mice might be due to impaired G-CSF signaling . To investigate the molecular mechanism that contributes to impaired G-CSF-dependent proliferation , we examined activation of STAT1 , ERK and STAT3 under the G-CSF signaling pathway ( Fig 6D ) . In repeated experiments , we found that the ratio of phosphorylated STAT1 , ERK and STAT3 vs . total STAT1 , ERK and STAT3 was markedly weaker and less prolonged in different level in MiR125a-/- neutrophils in response to G-CSF ( Fig 6E ) . This result indicates that the upstream in G-CSF signaling is impaired . However , we noticed that phospho-STAT3 was moderately enhanced while total STAT3 was much higher in MiR125a-/- bone marrow neutrophils . To determine whether the moderately enhanced p-STAT3 involves in mediating the decreased cell proliferation during maturation of MiR125a-/- GMPs , we cultured MiR125a-/- GMPs with G-CSF in the presence of STAT3 inhibitor S3I-201 or DMSO in CFU assays . Results show that inhibiting STAT3 cannot rescue the decelerated cell proliferation of MiR125a-/- GMP ( S5A–S5C Fig ) . Therefore , according to these data , STAT3 is unlikely to mediate decreased granulocyte differentiation in MiR125a-/- mice . Due to impaired G-CSF signaling pathway in MiR125a-deficient mice , we deduced that miR-125a might target a repressor in this signaling . SOCS3 is the major suppressor of G-CSF signaling and neutrophils differentiation [10 , 30 , 31] . Furthermore , we indeed detected higher SOCS3 protein expression levels in purified neutrophils lacking miR-125a compared to wild-type ( Fig 7A ) . Thus we tested whether miR-125a directly targeted Socs3 . We firstly predicted possible target sites in 3’UTR of Socs3 by using RNAhybrid and RNA22 , and we found miR-125a has a potential binding site in the 3’UTR of Socs3 ( Fig 7B ) . Then to confirm whether Socs3 is targeted by miR-125a , we cloned the full length of the 3’UTR of Socs3 onto a construct fused to the renilla reporter gene and mutated the predicted seed sequences . We co-transfected these plasmids with synthetic miR-125a oligonucleotide or negative control oligonucleotide in 293T cells respectively . The results indicated that miR-125a suppressed renilla luciferase activity but the mutants completely inhibited the suppression of the renilla luciferase activity ( Fig 7C ) . These results demonstrate that miR-125a directly targets Socs3 . But there remains a question whether Socs3 is a true target of miR-125a to regualte granulopoiesis . To address this issue , we did rescue experiments as follows . Firstly , we used shRNA to knock down Socs3 expression in MiR125a-deficient bone marrow cells and then did CFU assays in the presence of G-CSF . Results are shown that knockdown of Socs3 decrease Socs3 mRNA expression ( Fig 7D ) . And the colony size ( Fig 7E ) and the cell number per colony ( Fig 7G ) both increase after Socs3 knockdown . However , the colony number does not change ( Fig 7F ) . Next , we did a in vivo rescue assay by isolating short-term hematopoietic stem cells ( ST-HSCs ) from the bone marrow of MiR125a knockout mice , and we transduced these ST-HSCs with concentrated lentivirus of a Socs3 shRNA or a Ctrl shRNA , both of which contain GFP reporters . Then the transduced cells were collected and injected into the irradiated recipient wild-type mice . Six weeks later , the number of granulocytic progenitors and mature neutrophils was measured by FACS . Consistently with the results of in vitro CFU assay , we found that mice transduced with Socs3 shRNAs had significantly more GFP+ bone marrow neutrophils than those transduced with Ctrl shRNAs ( Fig 8A ) . However , the number of GFP+ granulocytic progenitor CMPs and GMPs was not affected after Socs3 inhibition ( Fig 8B ) . Furthermore in vivo BrdU-pulsing assays showed that BrdU incorporation of GFP+ CMPs and GFP+ GMPs did not change after Socs3 knockdown ( Fig 8C ) . Taken together , both in vitro and in vivo experiments successfully rescue the decelerated neutrophil development caused by miR-125a deficiency and further confirm that Socs3 is the main factor of regulating neutrophil development from GMPs to mature neutrophils rather than earlier progenitors in MiR125a deficient hematopoiesis .
Previous studies have demonstrated that ectopic expression of miR-125a contributes to expansion of hematopoietic stem cell pool [32 , 33] . However , we found an unexpected observation that the numbers of other mature hematopoietic lineage cells were not affected besides neutrophils in MiR125a knockout mice ( Table 1 ) . These inconsistent results might be explained by the reason that over-expression experiments may lead to gain-of-function phenotypes which cannot be found in knockout mice . Therefore , our results show miR-125a has an indispensable role in regulating neutrophil production . Neutrophils as well as monocytes-macrophages are the first line of defense in response to systemic inflammation caused by pathogen infection or injury . Under endotoxin challenge , monocytes-macrophages release inflammatory factors such as TNF-α recruiting neutrophils in several organs to mediate tissue destruction [26] . Depletion of neutrophils protects the liver against injury from endotoxin [34] . Thus , like monocyte-macrophages , neutrophils also play a crucial role in endotoxemia . Our study reveals that MiR125a-/- mice have decreased numbers of neutrophils compared to wild-type mice . In addition , in our LPS shock model , we observed resistance to a lethal dose of LPS in MiR125a-/- mice but the concentration of TNF-α and IL-6 in the serum remained unchanged compared to control mice . Furthermore , macrophage reconstitution experiments indicated that macrophages did not contribute to resistance to LPS shock in MiR125a-/- mice . Therefore , we eliminated the possibility that MiR125a-/- macrophages exhibited less cytokine production in response to stimulation of Toll-like receptors . Importantly , we found less neutrophil infiltration in the lungs and alleviated multiple organ damage in MiR125a-/- mice after LPS challenge . As we also detected MiR125a-/- neutrophils were as mature and functional as those in wild-type mice . Therefore , we deduced that resistance to a lethal dose of LPS in MiR125a-/- mice was mainly due to reduced neutrophil numbers in granulopoiesis . Granulopoiesis is part of hematopoiesis that maintains the peripheral neutrophil pool steady . In our study , we found MiR125a knockout mice showed neutropenia . We considered the main reason for the neutropenia was probably due to decreased cell proliferation from granulocyte progenitors to mature neutrophils in MiR125a-/- mice . The following are the main evidences demonstrated in this paper . Firstly , numbers of myeloid progenitors including CMPs and GMPs do not change according to FACS and CFU analyses , suggesting miR-125a may not regulate GMPs or even earlier progenitors . Secondly , the colony size is smaller and the cell number per colony is decreased from MiR125a-deficient GMPs , implying miR-125a involves in the respectively late stage of granulocyte development . Thirdly , immature and mature neutrophils are incorporated less BrdU in MiR125a KO mice while BrdU-incorporating promyelocytes/myelocytes have no change , meaning that miR-125a mediates cell proliferation during the differentiation from immature neutrophils to mature neutrophils . In addition , there is no difference in the rate of cell death between MiR125a-/- and wild-type mice by staining with Annexin V and propidium iodide , excluding the possibility that miR-125a-mediated cell death of neutrophils . Furthermore , other granulocytes ( eosinophils and basophils ) are not affected in MiR125a knockout mice ( Table 1 ) also indicating that miR-125a is specific for regulating immature neutrophils rather than affecting earlier common granulocyte precursors . To investigate the molecular mechanism of miR-125a in regulating neutrophil development , we checked the activation of G-CSF signaling pathway in wild-type and MiR125a deficient neutrophils . G-CSF is the major growth factor during each developmental stage of granulopoiesis [35] . STAT3 , STAT1 and ERK are downstream transcription factors in G-CSF signaling [36] . From western blot analysis , we found MiR125a deficiency mainly caused impaired G-CSF signaling pathway through weakening the phosphorylation ratio of downstream transcription factors . But it made us a little bit confused . Although the phosphorylation ratio of STAT3 was reduced , phospho-STAT3 was moderately enhanced while total STAT3 was much higher in MiR125a-/- neutrophils . In order to solve this problem , we used STAT3 inhibitor S3I-201 in GMP CFU assays . Results demonstrated that inhibiting STAT3 cannot rescue the decelerated differentiation from MiR125a-/- GMP . Thus we deduce that the phenomenon of the enhanced total STAT3 might be through other unknown mechanisms and it is unlikely to mediate decreased granulocyte differentiation in MiR125a-/- mice . Owing to the weak G-CSF signaling in MiR125a-deficient mice , we deduce that miR-125a might target a repressor in this pathway . SOCS3 is the principal suppressor of G-CSF signaling . It can bind to pY729 of the G-CSF receptor and directly inhibit receptor binding to JAKs , thus repressing downstream signaling [30 , 31 , 37] . Particularly the mice in which Socs3 is conditionally knocked-out in bone marrow have increased neutrophil number and enhanced cellular responses to G-CSF including an increase in proliferative capacity [10 , 11] . In our study , we actually identified Socs3 as a direct target of miR-125a . And the expression of Socs3 was indeed enhanced in MiR125a-/- neutrophils , weakening G-CSF signaling and eventually reducing neutrophils differentiation ( S6 Fig ) . Furthermore , both in vivo and in vitro rescue experiments demonstrated that Socs3 indeed was the main target of miR-125a to regulate late stage development of neutrophils rather than earlier progenitors . Nevertheless , we deduce that miR-125a promotes granulopoiesis mainly by targeting suppressor Socs3 . MiRNAs are abundant regulators of transcriptional programs . They serve as fine-tuners of biological systems by giving signaling pathways a threshold to protect from unwanted or wrong signals and making signal output more precise and appropriate [38] . In many signaling pathways , the expression of miRNAs can be induced or repressed in response to outside stimuli and form feed-forward or feedback mechanisms with other signaling components [13] . However , basal expression of miRNAs is important for cell-type-specific gene expression through acting as switches like transcriptional factors during cell lineage determination [39] . Hematopoietic lineage differentiation is also switched by miRNAs . For example , miR-150 for B cell [18–20] , megakaryocytic and erythrocytic lineage commitment [21] , and miR-223 for granulocytic differentiation [22 , 23] . In this paper , we proposed a model that miR-125a served as a positive regulator of physiological granulopoiesis by amplifying G-CSF signal strength and duration . In order to get a view of the regulation of miR-125a , we examined whether the expression of miR-125a was also affected by G-CSF signaling . However , we did not detect a significant change of the expression of miR-125a in granulocytes after G-CSF stimuli . As we found that miR-125a was decreased during maturation of granulocytes , we detected the expression of its target Socs3 which was also down-regulated and the expression of miR-125a and its target Socs3 exhibited a positive correlation in granulocyte development ( S7 Fig ) . Although this kind of correlation between miRNA and its targets is against the repressive nature of miRNA-mediated gene regulation , bioinformative analysis shows that it is prevalent [40] . Because miRNAs often repress target genes through translational inhibition and have minor effects on target mRNA levels , so miRNAs and their targets levels are mainly controlled by upstream transcription factors [40] . According to this model , both Socs3 and miR-125a are down-regulated during granulopoiesis and down-regulated miR-125a leads to up-regulated Socs3 as a feed-forward signal . Thus this circuit can tune upstream signal fluctuation and eventually maintain SOCS3 protein homeostasis . As Socs3 is a critical negative regulator of granulopoiesis , its level in progenitors of granulocytes can affect the neutrophils differentiation and any significant change may lead to pathological consequences , namely neutrophilia and neutropenia . From this view , miR-125a modulation eventually provides a steady device to maintain differentiation and homeostasis of neutrophils rather than to simply repress the expression of Socs3 . In conclusion , we showed that miR-125a can positively regulate granulopoiesis . We demonstrated that miR-125a positively regulated G-CSF-dependent proliferation during the development of granulocytes by targeting Socs3 . Our findings reveal a new microRNA involving granulocyte development and provide insights into the function of miR-125a during hematopoiesis . Future genetic studies will focus on how miR-125a is regulated during hematopoietic development .
MiR125a knockout mice were generated as previously described [25] and maintained under specific pathogen–free conditions at Institute of Health Sciences , Chinese Academy of Sciences animal breeding facility , according to institute guidelines . 8 to 12-week-old MiR125a knockout mice and their littermate controls were used for experiments . All experiments involving mice were in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals of 1988 , issued by the State Scientific and Technological Commission for China . And these experiments were approved by the Biomedical Research Ethics Committee of the Shanghai Institutes for Biological Sciences , Chinese Academy of Sciences . To analyze neutrophils , single cell suspensions of bone marrow or peripheral blood or spleen were stained with CD11b PerCP-Cyanine5 . 5 ( eBioscience 45-0112-82 ) and Ly-6G-APC ( eBioscience 17-5931-82 ) . To measure neutrophil infiltration in the lung , lung tissues were cut into very small fragments and digested by collagenase and DNase I for 20 minutes at 37°C . Single cell suspensions were then stained with CD45-FITC ( BD pharmingen , 553080 ) , Ly-6G-APC and CD11b PerCP-Cy5 . 5 . To detect the myeloid progenitor cells , bone marrow cells were pre-stained with biotin-conjugated mouse lineage panel ( BD pharmingen , 559971 ) , and then stained with streptavidin-V450 ( BD horizon , 560797 ) , Sca-1-PE-Cy7 ( BD pharmingen , 558162 ) , c-Kit-PE ( BD pharmingen , 553355 ) , CD34-FITC ( BD pharmingen , 560238 ) and CD16/32-APC ( eBioscience , 17-0161-82 ) . Flow cytometry was conducted on a FACS Aria ( BD Biosciences ) . The recipient mice were fed with acidic ( pH 2 . 6 ) , antibiotic water for one week before irradiation and then were given 8 . 0 Gy irradiation by using a 137Cesium Gammacell source . 4 hours later , the mice were injected with 2x107 bone marrow cells from the donor mice via tail vein and then were kept on giving acidic antibiotic water for the rest of their lives . To sort hematopoietic stem cells and progenitor cells , bone marrow cells were pre-enriched by depleting lineage positive cells ( Stemcell , 19756 ) . Hematopoietic stem cells were then sorted by Sca1+c-Kit+Lin- . CMPs were sorted by Sca1-c-kit+Lin-CD34+CD16/32- . GMPs were sorted by Sca1-c-kit+Lin-CD34+CD16/32+ and MEPs were sorted by Sca1-c-kit+Lin-CD34-CD16/32- . The purity of each cell population reached 95% . Neutrophils were isolated from bone marrow or peritoneal cavity by using the Neutrophil Isolation Kit ( Miltenyi Biotec , 130-097-658 ) . The purity of the isolated neutrophils was about 90% , as determined by flow cytometry . Total RNA was isolated using TRIzol reagent ( Life technologies ) . RNA quality was assessed with an Agilent 2100 Bioanalyzer ( Agilent ) , and only samples with an RNA integrity number > 8 were used . Global mRNA expression in bone marrow neutrophils with or without LPS stimulation samples from and MiR125a+/+ and MiR125a-/- mice were assayed with the Affymetrix GeneChip Mouse Genome 430 2 . 0 Array . Data were deposited in GEO ( GSE63739 , http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE63739 ) and analyzed with R and the associated BioConductor packages . Isolated bone marrow neutrophils were resuspended in 0 . 1% BSA 1X Hanks balanced salt solution containing calcium and magnesium ( Gibco ) and plated in 3 μm Transwells ( 1X105 cells per Transwell , Corning ) in the absence or presence of the indicated chemokine in the lower chamber ( 0 . 1 mM fMLP , Sigma; 250 ng/mL MIP-2 , Peprotech; 1μg/mL KC , Peprotech ) . After incubation at 37°C for 3 hours , numbers of cell that migrated through transwell were counted . Isolated bone marrow neutrophils were incubated in the presence of 1 μM dihydrorhodamine ( Sigma ) during stimulation with different concentrations of PMA ( Sigma ) for 15 minutes or LPS for 4 hours ( Sigma ) . The oxidative burst of neutrophils was then analyzed by flow cytometry . 2x105 Candida albicans strain SC5314 or 1x107 Citrobacter rodentium were incubated with or without 5x105 bone marrow neutrophils in flat-bottom 96-well plates for 4 hours . Then all wells were treated with 0 . 02% triton-X 100 in PBS for 5 minutes . Surviving bacteria or fungi were incubated with 10μl MTT ( 5mg/mL ) for 4 hours at 37°C then formazans were dissolved in DMSO and fluorescence was measured at 570 nm absorption wavelength . Total RNA was isolated with TRIzol reagent ( Life Technology ) . Expression of microRNAs in sorted cell populations was determined by quantitative PCR using the TaqMan MicroRNA Assay ( Applied Biosystems ) . MicroRNA expression was normalized to snoRNA202 . Socs3 , Il6 , Tnfa , Gcsfr , Pu . 1 , Gata-1 , Cebpa , Cebpb and Cebpe mRNA expression levels were quantified by using SYBR PrimeScript reverse-transcription–PCR kit ( Takara ) . Expression levels were normalized to endogenous expression of Gapdh . Wild-type mice were first depleted of endogenous macrophages by pre-treatment with 100 μl clodronate liposome Clophosome-A ( FormuMax Scientific ) on Day1 and Day2 . On Day3 , these mice were transplanted with 1x107 MiR125a +/+ or MiR125a -/- bone marrow-derived macrophages . Macrophage depletion was detected by flow cytometry on Day3 and Day6 and the spleen and bone marrow macrophages were depleted >90% . To count the number of GMPs , 5x104 bone marrow cells were cultured in methylcellulose ( Mouse Methylcellulose Base Medium , R&D Technologies ) added to various concentrations of recombinant murine G-CSF ( R&D Technologies ) . After 10 days , colony numbers were counted . To quantify multi-potential progenitors and lineage-restricted progenitors , 2x104 bone marrow cells were plated in complete methylcellulose medium ( Stemcell , 03434 ) . After 12 days , colonies were counted and analyzed morphologically . Bone marrow cells were cultured in 10% FBS RPMI 1640 medium ( Life Technologies ) for 48 hours , washed and stained for Ly-6G-APC and Annexin V FITC and PI ( BD Biosciences ) and analyzed by flow cytometry . For the in vivo BrdU-incorporation experiment , mice aged 8–10 weeks were intraperitoneally injected with 200 μl of a 10mg/mL BrdU solution . After 3 days , mice were sacrificed and the spleen and bone marrow cells were harvested to detect BrdU-positive neutrophils . For in vitro BrdU-labeling of cells , bone marrow neutrophils were isolated and stimulated with 100 ng/ml G-CSF for 24 hours followed by incubating cells with 10 μM BrdU for 1 hour . BrdU-positive neutrophils were detected by using the BrdU flow kit from Pharmingen ( BD Biosciences , 559619 ) with a FITC-labeled anti-BrdU antibody . Neutrophils were stained with CD11b-Percp Cy5 . 5 and Ly-6G-APC before fixation and permeabilization of the cells . TNF-α and IL-6 in mice serum were detected by R&D Technologies duo set ELISA kit . For immunoblotting experiments , bone marrow cells or neutrophils were lysed with RIPA buffer and blotted with indicated antibodies . P-STAT3 , STAT3 , p-STAT1 , STAT1 , p-ERK , ERK and SOCS3 were all purchased from Cell Signaling Technology . GAPDH antibodies were obtained from Abcam . To test whether miR-125a directly target the Socs3 3′ UTR , 293T cells were plated in 96-well plates and transfected with 10 ng wild-type or mutant Socs3 3′ UTR and the synthetic miR-125a oligonucleotide or negative control oligonucleotide by using Lipofectamine 2000 reagent ( Invitrogen ) . Firefly and renilla luciferase activities were determined after 24 hours using the Dual-Luciferase Reporter Assay System ( Promega ) . The values were normalized to firefly luciferase . To generate a retrovirus construct , MSCV-LTR miR30-PIG ( LMP ) plasmids were cloned into Socs3-specific hairpin RNA . The target sequence is as follows: CGC GAG TAC CAG CTG GTG GTG A . Plate-E cells were transfected with 30ug LMP shRNA for a dish and retroviruses were harvested from culture supernatant after 48 hours . Mice bone marrow cells were depleted lineage positive cells by magnetic beads , stimulated with G-CSF overnight , then infected with recombinant retrovirus . 48 hours later , green fluorescent protein expressing GMPs were sorted for CFU assays . To generate a lentivirus construct , pLVX-shRNA2 plasmids were cloned into Socs3-specific hairpin RNA . The target sequence is the same as above . 293T cells were transfected with 15ug pLVX-shRNA2 together with 8ug pMD2 . G and 15ug psPAX2 plasmids for one dish . Lentivirus were harvested and concentrated from culture supernatant after 72 hours . Bone marrow cells of MiR125a ko mice were depleted lineage positive cells by magnetic beads , and short-term hematopoietic stem cells ( ST-HSC ) were sorted by sca-1+c-kit+CD135-CD34+ and resuspended at 1 x 104 in 75 uL StemSpan ( StemCell Technologies ) , supplemented with 50 ng/ml SCF ( Peprotech ) in a round-bottomed well of a 96-well plate . 2 . 5 x 107 units of lentivirus were added into each well after 2 hours , predetermined to give about 20% transduction efficiency by measuring of GFP positive cells in pilot experiments . Then plates were spun at 900g for 90 min , and cultured at 37°C with 5% CO2-in-air . Cells were collected and washed after 4 . 5 hours and per 1 . 5 x 10^4 ST-HSCs were resuspended in 250 ul PBS which was then injected into each irradiated recipient wild-type mouse . | MicroRNAs are critical epigenetic modulators in development , physiology and disease processes . Many miRNAs are involved in immune cell development and function , like miR-150 for B cells , miR-181a for T cells . However , studies of miRNAs involvement in granulocyte development and function and related diseases are still limited . In this study , we developed engineering MiR125a knockout mice to study the function of miR-125a in vivo . We identified MiR125a knockout mice had decreased neutrophil numbers and reduced infiltration of neutrophils in the lung in LPS shock model . We deduced that this significant reduction of neutrophils was due to impaired development of granulocyte precursors to mature neutrophils in an intrinsic manner . Furthermore , we demonstrated that Socs3 , a major repressor that negatively regulates granulocyte development , was a target of miR-125a . This finding not only reveals a new microRNA involving granulocyte development , but also provides insights into the new mechanism of miR-125a during action in endotoxemia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"flow",
"cytometry",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"gene",
"regulation",
"immunology",
"bone",
"marrow",
"cells",
"animal",
"models",
"micrornas",
"model",
"organisms",
"granulocytes",
"experimental",
"organism",
"systems",
"neutrophils",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"gene",
"expression",
"mouse",
"models",
"stat",
"signaling",
"spectrophotometry",
"biochemistry",
"rna",
"cytophotometry",
"signal",
"transduction",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"non-coding",
"rna",
"cell",
"signaling",
"spectrum",
"analysis",
"techniques"
] | 2017 | MiR-125a Is a critical modulator for neutrophil development |
Sensory representations are not only sparse , but often overcomplete: coding units significantly outnumber the input units . For models of neural coding this overcompleteness poses a computational challenge for shaping the signal processing channels as well as for using the large and sparse representations in an efficient way . We argue that higher level overcompleteness becomes computationally tractable by imposing sparsity on synaptic activity and we also show that such structural sparsity can be facilitated by statistics based decomposition of the stimuli into typical and atypical parts prior to sparse coding . Typical parts represent large-scale correlations , thus they can be significantly compressed . Atypical parts , on the other hand , represent local features and are the subjects of actual sparse coding . When applied on natural images , our decomposition based sparse coding model can efficiently form overcomplete codes and both center-surround and oriented filters are obtained similar to those observed in the retina and the primary visual cortex , respectively . Therefore we hypothesize that the proposed computational architecture can be seen as a coherent functional model of the first stages of sensory coding in early vision .
In the last decades a large body of research has been devoted to explain the nature of neural representations . Since experimental manipulation of the stimuli has the most direct impact on the sensory responses , most of our knowledge comes from studies about the early stages of sensory systems . Although we do not have a complete story yet , experimental and theoretical research did reveal important principles about the nature of neuronal representations together with specific constraints imposed by anatomy and physiology . Derived from the efficient coding theory [1] , [2] , different popular models – emphasizing redundancy reduction ( like [3] , [4] ) or the sparsity constraint ( Sparse Coding , SC , e . g . [5] , [6] ) – can account for many , but not all relevant features of early sensory processing ( e . g . [7] , [8] ) . In this article we argue that a novel computational model of neural representation can be obtained by focusing on one of those relevant features: overcompleteness . For codes with this property the number of potential coding units is larger than that of the input units thus offering increased memory capacity and enhanced robustness against noise and structural perturbations . We will argue that the formation of large and sparse representations of high level of overcompleteness requires adaptive learning which can effectively control the number of active synapses . This structural sparsification has a significant impact on the overall metabolic cost of neural activity . We then present a new sparse coding scheme which is motivated by both theories mentioned above , but is built on a non-conventional signal model assuming an additive decomposition of stimuli into “typical” and “atypical” constituents . We also analyze the model's filtering properties when trained on natural images . The main contribution of our study is that principled pre-filtering based on this alternative signal model can indeed facilitate overcomplete SC by supporting structural sparsity . The pre-filtering process is motivated by recent results on efficient compression , completion and decomposition of high dimensional data; computational functions equally important for artificial and natural systems . Based on the finding that our model can simultaneously explain several features of early vision we then suggest a biological implementation of the two stage algorithm . The paper is organized as follows . In the Results section first we review the computational problem of overcomplete sparse coding and argue about the importance to control synaptic activity . Then we introduce our two stage algorithm which can achieve structural sparsity thus supporting overcomplete sparse coding . In support of our model numerical experiments on natural images are also presented . In the Discussion section we compare the computational properties and biological relevance of our model with alternative approaches . In the Methods section the details of the numerical experiments are provided together with brief descriptions , pseudocodes and references to more elaborate presentations of the algorithmic building blocks .
Due to the high metabolic cost of spiking activity [9]–[11] , constraining average spiking rate ( over time and population ) seems to be a general principle in neural systems ( but see [12] ) . Therefore we also consider sparsity central in our coding model . The objective of the sparse coding ( SC ) scheme is to find the sparsest representation of the data with low reconstruction error . It has been argued that this scheme offers a computationally and metabolically advantageous trade off between fully localized ( like “grandmother”-cells ) and distributed codes [13] . Sparse codes essentially try to approximate the underlying hidden structure ( the generating sources ) of the observed stimulus . The great advantage of SC over other coding schemes is that it directly controls energy consumption by setting the number of active coding units; out of coding units with can be active at any given time . Another important property of neural codes is overcompleteness , when the number of coding units ( ) is greater than the number of input units ( , ) . For example , in area 17 of cat the ratio of the output fibers versus the input fibers from the LGN is estimated about 25∶1 , while in macaque primary visual cortex , V1 the estimate is between 12∶1 and 160∶1 [14] or even 500∶1 [15] . In principle , overcompleteness provides more flexibility in finding even sparser representations . However , overcompleteness presents a non-trivial challenge for computational models on neural representations . In comparison with biological data , most computational models of SC can find the optimal solution if overcompleteness is 2 to 8-fold at most [16] . Importantly , higher level of overcompleteness may increase the overall metabolic cost of neural coding for two reasons . First , non-optimal solutions require too many iterations thus generating excess spiking activity . Second , overcompleteness induces an asymmetry in the use of the encoder and decoder channels within one iteration: while the excitation process requires the use of all encoder channels , selected subsets of active decoding units require only decoder channels . That is methods that avoid the heavy use of encoding are more favorable . The importance of controlling the number of active coding channels ( that is the number of synapses which define the receptive field of a neuron ) is highlighted by the fact that according to the estimates of [10] , more than 50% of the metabolic cost of a single spike can be attributed to the excitatory potentials at the postsynaptic sites ( EPSPs ) . Our goal is thus to find an algorithmic model that can explain overcomplete sparse coding in the brain . Formally , SC can be stated as an alternating ( two step ) optimization problem: ( 1 ) where is the signal , or input to be reconstructed , is the number of training inputs , ( ) denotes the coefficient vector of the sparse decomposition also called ( internal ) representation and is the basis , or dictionary of features . denotes the -norm , which is the number of nonzero components . The first term minimizes the reconstruction error , while the second one penalizes solutions with many non-zero components . Sparsity of representation is defined as where is the number of non-zero components . The resulting code is overcomplete , if and the difficulty of finding a sparse code with minimal reconstruction error depends on the level of overcompleteness ( ) and . Parameter controls the trade-off between the two terms . The reconstruction error or residual may be due to different noise sources that hide the structure of generating sources of the signal . At one step the basis set is adjusted ( learning process ) to minimize the reconstruction error while the activity of the coding units , is kept fixed . The straightforward solution would be to let evolve by stochastic gradient on the cost function derived from the reconstruction error , where and ‘hat’ denotes the actual estimation . Because of the role of the reconstruction error , this rule is not directly local [17] , yet it can be translated [18] into a set of Hebbian ( local ) interactions realized by particular network structures with feedback . During the selection of non-zero units ( formation of the sparse code ) , features ( ) are fixed . However , selection by exhaustive search is a combinatorially hard problem [19]: the number of iterations becomes computationally prohibitive as ( the dimension of the internal representation ) increases . For this reason several approximation method exist , but they either have slow convergence or provide non-optimal solutions . To overcome these limitations , we have chosen a heuristics that combines two approaches . The so called Subspace Pursuit ( SP ) method [20]–[23] has been chosen because of its superior speed . It is a generalization of matching pursuit [24] , which finds local optima in a fast iterative fashion . Importantly , this method is able to discover the global optimum provided that certain conditions are met . Numerical experiments on natural visual stimuli indicate that methods , which assume these conditions , work surprisingly well [25] , even though the conditions are unlikely to be met ( but see [26] on the inherent limitations of matching pursuit like methods ) . In contrast to SP , the other algorithmic component – the so called Cross Entropy method ( CEM ) [27] – is an optimization method designed to find the global optimum . Its main limitation is the slow convergence rate . The combination , termed Subspace Cross-Entropy ( SCE ) [16] method inherits the best of both worlds: it is reasonably fast and still can yield the optimal solution even at a higher level of overcompleteness . Since we are interested in the formation of sparse codes at very high level of overcompleteness , we used SCE in our numerical experiments . The appendix contains the pseudocodes of SP , CEM and SCE for the sake of reproducibility . Detailed analysis of these methods can be found in [16] , [28] , [29] . The learning process of Eq . ( 1 ) is prune to perturbations: excess activation caused by noise may induce changes in all features thus introducing global ( long-range ) and low spatial frequency correlations among the features . Such unwanted increase in the number of active synapses implies increased metabolic cost . Observation noise ( e . g . induced by intrinsic neural activity ) can significantly decrease the efficiency of OSC as it may easily generate access activation at the output ( representation ) level , which can only be mitigated by a number of further iterations in order to reduce the reconstruction error . In turn it is essential to counter this effect by actively controlling the number of non-zero components of the filters . This constraint is referred to as structural sparsity and implies that visual RFs with local , i . e . , spatially restricted responses ( like the high frequency , concentric RFs of the retinal ganglion cells , the relay neurons in the LGN , or the elongated oriented Gabor patch like RFs of the simple cells in V1 ) are metabolically more favorable over those that have large global structure with many synapses involved [30] . Approaches like weight thresholding or increasing overcompleteness ( see Discussion ) fail to address this issue properly . Instead , we turn to an alternative approach by directly separating global ( involving many synapses ) , i . e . , low-frequency or long-range components of the stimuli before the actual sparse coding . Considering the famous frequency fall of the amplitude spectrum of natural images [31] , the low-frequency components carry most of the energy . Principal Component Analysis ( PCA , [32] , often used decorrelation method ) , for example , represents the signal in a way that the first component would carry the largest amount of energy , while the last one would carry the least amount . In turn , by applying PCA and then projecting the data out of the subspace of the first principal components would yield a representation without the unwanted low-frequency content . Let us remark that this approach is in contrast to conventional thinking which would keep exactly those components with high energy and filter out the rest . While this idea is appealing , PCA based separation of the subspaces strongly depends on the signal statistics: components ( “outliers” ) with heavy tailed amplitude distribution ( characteristic to natural stimuli ) can easily break down PCA . In the next section we review a robust alternative to PCA , which can efficiently separate these outliers from the low frequency components . We then propose an overcomplete SC model in which SCE ( or any other efficient SC solution ) is complemented by this alternative prefiltering as it is expected to support structural sparsity in the subsequent SC stage . Our concept is based on recent findings of signal processing about recovering low-dimensional data from high dimensional observations [33] . In signal processing , conventional analysis of large dimensional data , such as sensory observations , is often based on the assumption that data have low intrinsic dimensionality: they lie on a low-dimensional subspace . In norm ( the -norm of vector , where stands for transposition , is defined as ) , PCA provides rank- estimate of the data by solving the following problem: ( 2 ) ( 3 ) ( 4 ) where is the matrix of observations ( dimension of the observations: , number of data points: ) , rank of matrix is at most and models a small noisy perturbation of each entry . If this perturbation is Gaussian noise , then PCA provides the statistically optimal estimate of the low-frequency , low dimensional subspace . However , deviation from the Gaussian ( e . g . gross perturbations or components with heavy tailed distribution ) can easily yield incorrect estimates . Because of the frequency dependence natural stimuli often contain outliers and thus we need an alternative signal model . Let matrix comprise the low frequency components ( so it has low-rank as above ) , while may have full rank , but it is a sparse matrix with arbitrarily large entries at random locations: . The surprising result is that under certain conditions ( on the rank of and on the sparsity of ) both matrices can be exactly recovered [33] . Furthermore , it has been proved that efficient recovery is feasible by solving the following optimization problem ( Robust Principal Component Analysis , RPCA ) : ( 5 ) ( 6 ) where denotes the sum of the singular values of , denotes the norm of matrix , i . e . , . is a trade-off parameter , which governs the dimension of matrix . On the other hand , matrix may assume maximal rank , independent of . In addition to robustness against perturbation , the proposed decomposition allows an alternative interpretation of the signals . Instead of treating sparse components as corrupting noise to be filtered , we may consider these outliers as atypical signals that carry further information about higher order correlations ( like configurational information ) not revealed by the low-rank estimate ( ) . Note that conventional methods ( like ICA ) would analyze the low rank part only . The suggested solution ( the pseudocode is given in Table 1 ) iteratively improves the estimation of and and its computational complexity is only slightly larger than that of the traditional PCA [33] . Another surprising result is that under the assumptions of the theorem , a whole range of values can return the correct solution , no matter what and are . A simple reference value for is [33] and so we will use a normalized parameter: . Interestingly , as numerical experiments suggest [33] , RPCA delivers meaningful signal decomposition even if conditions ( about the sparseness of ) do not hold ( like in the case of spectra ) . In these cases , however , different RPCA decompositions can be obtained by setting different values and is not guaranteed to be sparse anymore . For this reason matrix could be the subject of further sparsification . The corresponding sparse coding optimization ( see Eq . ( 1 ) ) in matrix form is given as ( 7 ) where the matrix and , denotes the matrix of the outliers and the matrix of their sparse representations , respectively . The norm based residual may denote full rank observation noise , which implies the following signal model: . According to [34] , it is still possible to give stable estimates for and , if is bounded: , for some value , where denotes the Froebenius norm . In the demonstrations we opted to use the simpler RPCA model ( as in Eq . ( 6 ) ) without explicit assumptions about the additive noise term . Let us note that even though the formalism used above is based on matrices , the RPCA procedure can be applied on a single input ( thus it may be realized in a neurally plausible form ) once an approximation of the low-rank part is available . Furthermore , – depending on the input statistics – can be approximated even from partial observation by ‘filling in’ missing information [33] , [35] . To test the impact of RPCA preprocessing on sparse coding , normalized natural image patches were first decomposed by RPCA at different values , then the resulting full rank representations were further encoded by SCE ( 16-fold overcompleteness with dimensional inputs and dimensional representation; numerical details are in the Methods Section ) . We have chosen this particular input set since there already exist a number of computer vision studies on their statistics and the corresponding neural representations under different optimality criteria [14] , [31] . The actual overcomplete sparse representations were formed by SCE and the corresponding SC filters were tuned online via stochastic gradient learning . While this level of overcompleteness is still below what has been estimated in the neural sensory systems [15] , we believe it is a reasonable choice , as training time is still manageable , yet the results are convincing enough to support the central message of our proposal . A few basis features ( for sparse coding , 10 out of 4096 columns of matrix ) are shown on Figure 1 . For visualization purposes each basis vector is scaled into the range and displayed as a image . Features in the first row of Figure 1A were obtained by conventional SC ( applying SCE ) without pre-filtering , which corresponds to the case of . As we earlier argued , plain SC tends to learn large , global filters , thus preventing the reduction of synaptic cost . Figure 1B plots a few selected SC features when applied on the residuals of traditional PCA . Regarding locality we do not see much improvement: features are still global and manifest large , wavy structures . Figure 1E depicts example filters obtained by applying RPCA prior to SC . Different rows correspond to different values . The main result of these studies is that the learned basis features get cleaner and more localized , that is , filters get structurally sparser as the single global parameter increases . On Figure 1F we re-plotted features for together with the corresponding filters approximated by reverse correlation . Not only the estimation error is smaller compared to the error of the native SC method ( Figure 1A ) , but filters also show larger diversity in their shapes , similar to what has been found experimentally [8] . We also plotted the corresponding filters or RFs of the low-rank signal in Figure 1C for , when the number of basis vectors was 17 . Figure 1D shows the spatial-dependence of RFs of the sparsified signal after RPCA for . A surprising result is that the shape of all the obtained RFs for sparsified matrix can be described as ‘Difference of Gaussians’ which is the characteristic RF shape [36] of the retinal ganglion cells and the neurons in LGN . The obtained concentric filters 1 , are homogeneous and 2 , uniformly tile the whole space . Due to their similarity , we show the cross-section of one unit only ( Figure 1D ) . Note that the peaky structure is due to the small image size ( discretized DoGs have similar shape at this scale ) and more typical DoG shapes could be obtained for larger image patches . We found that for higher values the negative basin around the peak gets deeper . This development may correspond to the experimentally found developmental changes of the LGN filter profiles in cat [37] . Let us emphasize again that RPCA is not a projection: through an iterative process it extracts the large and sparse components and separates the low-rank part . Interestingly , for natural images , RPCA provides a basis visually almost indistinguishable from those of the PCA filters , but the corresponding representations are different . It implies that PCA may be a good first approximation or initialization for the RPCA iteration method ( higher values allow more low-dimensional components ) . Traditionally , a simple cell RF in V1 is often characterized as a ‘Gabor-patch’ [38]; Gaussian envelope around a cosine wave . To help compare the obtained filters with RFs of real neurons , we also approximated the filters as a Gabor-patch . As increases the filters become more localized and cleaner , and the Gabor-patch like appearance gets more pronounced . On the other hand , at too large values the filters become small and stereotyped with diminishing harmonic content ( see Figure 1E ) . The distribution of the shape parameters of the Gabor-patch approximations ( Eqs . ( 9 ) – ( 11 ) ) is shown in Figure 2 for . Filters localized at the edges of the visual space were discarded as their distortion prevents proper fitting . For small filters fitting is imprecise . Filters yielding Gaussian envelope with width less then 0 . 3 pixel were thus also discarded . It implies that the true number of learned filters at around point is larger than what is shown in Figure 2 . Visual inspection reveals that ( i ) filters become local and cleaner , ( ii ) the distribution deviates significantly from the bisection line , and ( ii ) a considerable portion of the filters is concentrated near the origin . For comparison , we also plotted the distribution of the fitted shape parameters of the experimentally measured RFs of simple cells reproduced from [8] . Considering that we had to drop a number of small filters , the match between numerical and experimental data seems quite good ( see , e . g . , [39] for comparison ) , indicating that the proposed model may have biological relevance . Let us note that the observed shape distribution may depend on the level of overcompleteness , but due to the relatively small input size we suspect that further increase in the number of coding units would not result in major changes . Since the assumed signal model is only an approximation for natural image patches , different trade-offs ( defined by in Eq . ( 5 ) ) between the contribution of the typical and atypical features to the reconstruction influence the emerging representations after RPCA prefiltering . Figure 3A depicts the influence of and thus the RPCA decomposition on the statistics of the SC filter shapes as measured by the histogram of the Gabor-patch fitting error . It shows how well the linear approximation of sparse coding filters can be described with a set of oriented Gabor patches often used to characterize experimentally measured receptive fields . If filters have ‘dilated’ global structure then the histogram of the fitting error is probably less peaked . And indeed this is the case: increasing results in more homogeneous , smaller and point-like filters . Let us remark that discretization has a strong contribution to the observed fitting noise . Figure 3B displays the dependence of the dimension of the low-rank component on and the relative contribution of to the reconstruction of the original observations . To calculate the intrinsic dimension of , all singular values were zeroed out with amplitude less then of the maximal amplitude . The important parameter range is where the intrinsic dimension is still low , yet role in the reconstruction is significant . Within that range , provides the best fit to the experimental data . At higher values most of the filters loose their edge-like characteristics . We have also studied the algorithm's reconstruction ability . Due to the additive decomposition , reconstruction depends on both the “typical” part obtained by RPCA and the overcomplete sparse representation of the “atypical part” . As it is demonstrated on Figure 3 the relative contribution of as well as its dimension ( number of coding neurons ) depends on . In turn , the fidelity of reconstruction is a function of both the number of units that encode typical features and the number of nonzero entries in the sparse code . Figure 4 displays this dual dependence: reconstruction quality as a function of the total number of nonzero entries , which comprises the rank estimate of at the given and the preserved number of nonzero entries in the overcomplete sparse representation ( ) . For the chosen values were: and for , . Reconstruction quality is measured by mean SNR: . Interestingly , while SNR does not improve much when has changed from to , the corresponding filters have significantly changed . Let us note that the overall low values of SNR are due to the fact that no high frequency components have been filtered out prior to decomposition ( but see [40] , where much higher SNR has been reported after filtering out those high frequency components ) . So far we have dealt with static images , but temporal sequences are more realistic: sensory systems are believed to adapt to the spatio-temporal structure of the stimuli . Since RPCA does not rely on prior knowledge about the spatial or temporal arrangement of the data , one expects to see similar decomposition results for data with temporal correlation . For the sake of illustration , temporal correlation was introduced by concatenating 16 image patches of size 8×8 extracted from image sequences on natural scenes . ( This was the maximum size we could handle with overcompleteness ratio 16 . ) Sample filters of the obtained low-rank matrix for ( the corresponding rank estimate is ) are shown on the left of Figure 5 . Filters are ordered by their corresponding eigenvalues . Each filter is composed of 16 frames of size 8×8 pixels . Similar to the filters shown on ( Figure 1C ) , these filters can also be characterized by low spatial and temporal frequency . The corresponding filters of the atypical parts ( , not shown ) - as in the static case- are homogeneous , localized in space and time and uniformly tile the visual space . Furthermore , they show Mexican hat like characteristics in the temporal dimension . The regularity may be due to the particular concatenation method we chose . Sparse coding filters can also be derived from the overcomplete sparse representation of the image sequences after RPCA decomposition . As representations are temporally decorrelated , we obtained filters strongly localized in space and time which resemble to some extent to the receptive field dynamics of simple cells of V1 [41] . A sample set of the obtained sparse coding filters are shown on the right of Figure 5 . It is expected to get better match with experimentally found filters if temporal correlations are introduced into the data model by convolution [42] , [43] as opposed to simple concatenation and if nonlinear response properties and nonlinear dynamic interactions are included to handle time warping , for example . These studies go beyond our present goals .
The biological relevance of neural coding models is often judged by the similarity between their filtering properties and the receptive fields of the corresponding neurons . In the case of visual stimuli , one of the criticisms against theory driven ( functional ) models ( e . g . Independent Component Analysis [4] or Sparse Coding [5] ) is the lack of diversity in the filter shapes [8] . This failure might be due to the missing prefiltering stages as seen in the visual pathway . However , nave use of different , biologically motivated prefiltering methods does not seem to offer any improvement , either . For example , applying DoG as high-pass filtering is expected to enhance edge-like features thus yielding a shift of the Gabor-patch shape parameters toward higher values , but the structure of the shape distribution barely changes . Another example is the use of PCA to filter out global features before SC ( or ICA ) , which yields wavy SC basis ( Figure 1B ) . Furthermore , not all filters in V1 have elongated bar shape and most models fail to yield close to concentric shapes found experimentally ( for a discussion , see e . g . [39] ) . As the filter shape distribution on Figure 2 shows , when applied on natural images , RPCA preprocessing together with SC delivers the required diversity including the close to concentric shapes . It is worth noting there are other improved coding models ( in particular , [40] and [39] ) that also claim similarities between the observed and predicted shape distributions of the fitted filters . Our model is similar in spirit to the functional model of [40] , whereas the other approach [39] describes a self-organizing system governed by complex dynamics and feedforward inhibition . While the latter one is a promising approach , its dynamics is quite involved and its parameter sensitivity is not known . The other model of [40] is also a sparse coding model and it uses greedy , iterative solutions as mentioned previously . It also uses prefiltering similar to that one used in [5] . They claim the obtained similarity is due to the particular sparsity constraint . For the similar motivations let us remark some differences between the model of [40] and the one proposed here . First , we believe their approach may not be suited to handle large overcompleteness for reasons discussed previously about greedy solutions . Second , the reported difference between the signal to noise ratio of their method and our model is likely due to two factors: we did not employ prefiltering and the overcompleteness in our case is larger . Less sparse codes can encode signals more faithfully then . A fair comparison would be to see the quality of the reconstruction of the high frequency components from sparse codes ( ) , but such comparison would depend on both sparsity and overcompleteness . In turn , an intriguing issue is the optimality of reconstruction quality with respect to the energy consumption . Interestingly , as Figure 6 demonstrates , the linear approximation of the filtering properties of RPCA ( seen as the amplitude spectrum of the “atypical” signal part of the RPCA output ) looks quite similar to what an ideal whitening filter would yield . This similarity may have the following consequences . First , their result may be attributed both to the particular form of the filter and to the chosen form of sparse coding . Furthermore , it might be the case that such prefiltering behaves as a fast approximation to RPCA . Another difference to mention is that our two-stage model not only provides oriented band pass filters , but it also yields DoG-like filters at the RPCA pre-filtering stage thus providing a simultaneous explanation of two processing stages of early vision . Interestingly , as Figure 6 demonstrates , linear approximation of the filtering properties of RPCA ( seen as the amplitude spectrum of the “atypical” signal part of the RPCA output ) looks quite similar to what an ideal whitening filter would yield . This similarity may have the following consequences . First , results of [40] may be attributed both to the particular form of the filter and to the chosen form of sparse coding . Furthermore , it might be the case that such prefiltering behaves as a fast approximation to RPCA . The qualitative agreement between the filtering properties of the early stages of vision and our two-stage algorithm may allow us to attempt to map the algorithm onto the neural substrate by linking the different computational functions to anatomical areas . An important property of our model is that prefiltering requires a dual representation of the stimuli , which assumption is not in line with the current thinking of hierarchical sensory processing ( e . g . [45] , [46] ) , which often comprises alternating filter and pooling operations . So how can we reconcile the assumption on dual representation with single stream models ? Since RPCA implies dynamic interaction between the two emerging representations of the typical ( global ) and atypical ( local ) features , decomposition requires either a recurrent network with distinct sub-populations of neurons or two layers with feedforward and feedback connections . As retina does not receive feedback modulations from downstream layers , DoG like filtering of the retinal ganglion cells is not a consequence of RPCA , but it may be explained as a facilitating approximation – as we argued about whitening above – before decomposition . LGN , on the other hand , receives massive amount of feedback from V1 . Having learned the filters during early development , it can be assumed that LGN neurons can represent a proxy to the atypical features of single stimuli . This representation still contains information about the typical features ( since clear decomposition of natural signals is unlikely , due to scale-free statistics ) . In turn , V1 has a two-fold role in processing . It holds the approximation of the global features extracted from the LGN output and it recodes or re-represents the atypical features in an overcomplete sparse form . A candidate for the first task could be a class of V1 interneurons characterized by large , global receptive fields with weak or no orientation selectivity ( e . g . [47] , [48] ) . While it is possible to learn the low-frequency typical parts of new stimulus sets , RFs do not need to be continuously updated as they comprise the most typical correlations of natural images ( short term adaptation to quick changes is still required ) . The second task of overcomplete recoding is then realized by simple cells . This setting thus allows for the alternating substraction of RPCA ( Table 1 ) by the interaction between inhibitory neurons and simple cells in V1 and the neurons in LGN . In summary , this paper presents a novel two-stage algorithm for efficient overcomplete sparse coding . The proposed robust extraction of low-frequency or typical correlations as a prefiltering step has a few remarkable properties that make the algorithm plausible as an important model of neural information processing . First , it supports the formation of overcomplete sparse codes by effectively controlling the transformation matrices ( the synaptic weights ) and reducing the number of active synapses . Second , the inclusion of RPCA could significantly facilitate perception as it allows the completion of the typical components even if a part of the stimuli is missing ( undersampling , occlusion , cf . exact matrix completion ) . Since these properties may be beneficial for the nervous system , it would be interesting to see if our algorithm could be realized by biologically plausible neural computations .
Subspace Pursuit algorithms have been independently proposed in [23] and [49] . These methods assume that at most components are sufficient to represent the input . The methods enlarge the subset of candidate features ( “candidate subspace” ) by [23] ( or [49] ) features and then decrease their number back to at every iteration . The method of [23] is as follows ( the pseudocode is given in Table 2 ) . First , a candidate representation is generated using all basis , then a subset of basis is selected that corresponds to the largest components of the representation . This initial selection is then iteratively refined: the residual ( that is the difference between the input and its current approximation ) is calculated and mapped onto the representation space using the entire basis set again . Then – similar to the initial step – another basis are selected based on amplitude of the corresponding components of the mapped residual . The original input is then projected again to the representation space using a element basis set formed by fusing the two basis subsets . Finally basis vectors are selected again that correspond to the largest components of the projection ( basis shrinkage ) . The iteration stops when the norm of the residual is sufficiently small . SP has superior speed , scaling and reconstruction accuracy over other iterative methods by directly refining the subset of reconstructing ( active ) components at each iteration . Its native shortcomings , though , are the heavy use of the costly encoding transformation of the residuals at each iteration and the preset number of active coding units . CEM is a global optimization technique [27] that finds the solution in the following form:where is a general objective function . While most optimization algorithms maintain a single candidate solution at each time step , CEM maintains a distribution over possible solutions . From this distribution , solution candidates are drawn at random . By continuous modification of the sampling distribution , random guess becomes a very efficient optimization method . One may start by drawing many samples from a fixed distribution and then selects the best samples as an estimation of the optimum . The efficiency of this random guess depends on the distribution from which the samples are drawn . After drawing a number of samples from distribution , we may not be able to give an acceptable approximation of , but we may still obtain a better sampling distribution . The basic idea of CEM is that it selects the best few samples , and modifies so that it becomes more similar to the empirical distribution of the selected samples . CEM resembles the estimation-of-distribution evolutionary methods ( see e . g . [50] ) and as a global optimization method , it provably converges to the optimal solution [27] , [50] . For many parameterized distribution families , the parameters of the minimum cross-entropy distribution can be computed easily from simple statistics of the elite samples . For sparse representations the Bernoulli distribution is of particular interest [51] . This particular choice may bring about bias towards solutions where sparse components are drawn independently . Derivations as well as a list of other discrete and continuous distributions with simple update rules can be found in [52] . Let us note that we have also translated CEM into an online variant in which parameter tuning is realized by neurally plausible local learning [29] . This translation then allowed us to propose a neurally plausible SC method [28] in which spikes signal the presence of active components , while rate codes encode the corresponding uncertainty of the given component . Since CEM randomly generates candidate sparse solutions hand , it uses a significantly less number of costly encoding transformations . However it updates the probability of all active components similarly , regardless their individual contributions to the actual reconstruction error . Subspace Cross-Entropy method ( SCE ) is an efficient combination of CEM and SP for overcomplete sparse coding . A detailed description can be found in [16] and the pseudocode is given in Table 3 . SCE inherits the flexibility and synaptic efficiency of CEM as well as the superior speed and scaling properties of SP without their shortcomings . SCE can be realized by inserting an intermediate control step in CEM to individually update the component probabilities based on their contribution to the reconstruction error . Hence the explicit refinement of the feature set via SP is replaced by an implicit modification through component probabilities . Since the resulting algorithm is not a greedy method , the algorithm is called as Subspace Cross Entropy ( SCE ) method without the term ‘Pursuit’ . An efficient implementation of RPCA algorithm rephrases the optimization problem of ( 5 ) by means of the augmented Lagrangian with the following objective function [33] ( 8 ) where denotes the current residual after subtracting and . The efficiency stems from the fact that both and subproblems have simple solutions . Let denote , which can be applied componentwise on matrices . For matrices , let denote the singular value thresholding operator , where is any singular value decomposition . The corresponding pseudocode is given in Table 1 . The algorithms were trained on 16×16 , normalized ( zero mean and 1 std ) patches extracted from a public database ( http://www . cis . hut . fi/projects/ica/data/images/ ) . For the temporal studies , inputs were generated by concatenating 16 normalized patches of size 8×8 extracted from randomly selected parts of publicly available videos ( ‘football ( b ) ’ , ‘garden’ , ‘ice’ , ‘tempete’ , ‘crowd_run’ , ‘sunflower’ , ‘tractor’; http://media . xiph . org/video/derf/ ) . To speed up calculations , batch learning ( 50000 samples for static stimuli and 25000 samples for the sequences ) was applied to learn the low dimensional subspace of RPCA in the preprocessing stage . On the other hand , to learn the over-complete sparse basis ( -fold over-completeness ) , samples have been used . RPCA was run in MATLAB . All other transformations were performed on a cluster of 17 Sony PlayStation 3 consoles in Linux environment using in-house C++ implementation of published algorithms of SVD [53] and CE [27] . The obtained filters were matched with Gabor filters [36] , [38] in order to characterize the spatial structures . The Gabor filter parameters are as follows: ( 9 ) ( 10 ) ( 11 ) where and denote the center of the patch , is the orientation of the normal to the parallel stripes of the Gabor function , is the frequency and is the phase of the cosine factor , and specify the ellipticity of the Gaussian envelope . Fitting was done in MATLAB using the nonlinear least squares optimization function ( nsqnonlin ( . ) ) designed for large scale problems . For each parameter value the optimization algorithm was run 20 times with random initialization and the best solution was kept . | Neural systems favor overcomplete sparse codes in which the number of potential output neurons may exceed the number of input neurons , but only a small subset of neurons become actually active . We argue that efficient use of such large dimensional overcomplete sparse codes requires structural sparsity by controlling the number of active synapses . Motivated by recent results in signal recovery , we introduce a particular signal decomposition as a pre-filtering stage prior to the actual sparse coding , which efficiently supports structural sparsity . In contrast to most models of sensory processing , we hypothesize that the observed transformations may actually realize parallel encoding of the stimuli into representations that describe typical and atypical parts . When trained on natural images , the resulting system can handle large , overcomplete representations and the learned transformations seem compatible with the various receptive fields characteristic to different stages of early vision . In particular , transformations realized by the prefiltering units can be approximated as ‘Difference-of-Gaussians’ filters , similar to the receptive fields of neurons in the retina and the LGN . In addition , sparse coding units have localized and oriented edge filters like the receptive fields of the simple cells in the primary visual cortex , V1 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"neuroscience",
"biology",
"sensory",
"systems",
"neuroscience"
] | 2012 | Efficient Sparse Coding in Early Sensory Processing: Lessons from Signal Recovery |
Trypanosoma evansi is the parasite causing surra , a form of trypanosomiasis in camels and other livestock , and a serious economic burden in Kenya and many other parts of the world . Trypanosoma evansi transmission can be sustained mechanically by tabanid and Stomoxys biting flies , whereas the closely related African trypanosomes T . brucei brucei and T . b . rhodesiense require cyclical development in tsetse flies ( genus Glossina ) for transmission . In this study , we investigated the evolutionary origins of T . evansi . We used 15 polymorphic microsatellites to quantify levels and patterns of genetic diversity among 41 T . evansi isolates and 66 isolates of T . b . brucei ( n = 51 ) and T . b . rhodesiense ( n = 15 ) , including many from Kenya , a region where T . evansi may have evolved from T . brucei . We found that T . evansi strains belong to at least two distinct T . brucei genetic units and contain genetic diversity that is similar to that in T . brucei strains . Results indicated that the 41 T . evansi isolates originated from multiple T . brucei strains from different genetic backgrounds , implying independent origins of T . evansi from T . brucei strains . This surprising finding further suggested that the acquisition of the ability of T . evansi to be transmitted mechanically , and thus the ability to escape the obligate link with the African tsetse fly vector , has occurred repeatedly . These findings , if confirmed , have epidemiological implications , as T . brucei strains from different genetic backgrounds can become either causative agents of a dangerous , cosmopolitan livestock disease or of a lethal human disease , like for T . b . rhodesiense .
Trypanosoma evansi is an important disease-causing parasite of livestock in many African , Asian and South American countries . T . evansi belongs to a group of five closely related named taxa of various ranks found in a wide diversity of mammalian hosts; Trypanosoma brucei brucei , T . b . gambiense , T . b . rhodesiense , T . evansi , and T . equiperdum [1–5] . Three of these closely related parasites ( T . b . brucei , T . b . gambiense , and T . b . rhodesiense ) are only found in sub-Saharan Africa , where they require transmission by a tsetse fly vector and cause nagana in animals and sleeping sickness in humans , respectively [6 , 7] . The other two members of this group ( T . evansi and T . equiperdum ) are found both inside and outside the African continent , use other means of transmission , and are responsible for surra in wild and domestic animals [8] and dourine in equines [9] , respectively . The formal taxonomy of this group of closely related trypanosomes is in flux and currently reflects their disease outcome and means of transmission rather than their evolutionary relationships [10–14] . For example , strains of the human infective named subspecies , T . b . rhodesiense , are genetically closer to different T . b . brucei strains than to other strains from the same named subspecies [3 , 13 , 14 , 15] . Similarly , the taxonomic rank of T . evansi and T . equiperdum is in question because the few T . evansi and T . equiperdum strains that have been analyzed to date are genetically closer to different T . b . brucei strains than to other strains from the same named species [1 , 10 , 11 , 12 , 13 , 14 , 16 , 17] . This indicates that neither named species is monophyletic and suggests multiple origins from T . b . brucei . Despite the clear need for taxonomic revisions , and to avoid confusion , we use the established nomenclature . We further classify T . evansi based on their mitochondrial DNA ( kinetoplast DNA or kDNA ) configuration of type A or B [18–20] and their antigenic variant surface glycoprotein ( VSG ) Rode Trypanozoon antigenic type ( RoTat ) 1 . 2 , used in serological and PCR-based diagnostic tests [11 , 21 , 22 , 23 , 24] . Trypanosoma evansi is the most geographically widespread of these trypanosomes [2 , 25] , and some authors have suggested that it originated in camels in Africa [8 , 12] , where it occurs in all countries where these animals are found . This distribution extends along a northern line from Senegal to Mauritania , Morocco , Algeria , Tunisia , Libya , Egypt , Sudan , Eritrea , and Ethiopia , and the northern parts of Mali , Burkina Faso , Niger , Nigeria , Chad , Somalia , and Kenya [8] . Outside of Africa , T . evansi is thought to be limited by dispersal routes rather than the presence of camels and occurs in Asia and South America [2] . Both inside and outside of Africa , surra affects a variety of animals besides camels , including horses , cattle , buffalos , small ruminants , and dogs [2 , 26] , causing thousands of animal deaths per year . Although the net economic losses attributable to T . evansi infections are difficult to estimate [2 , 26] , mortality rates of animals affected and total effort invested in chemotherapeutic interventions indicate significant economic losses and social impacts among regions of the world [5 , 27 , 28 , 29] . African trypanosomes within the T . brucei complex require cyclical development within the tsetse fly vector to complete their life cycle and transmission [30 , 31] . In contrast , T . evansi and T . equiperdum exist exclusively as developmental forms equivalent to the bloodstream form of T . brucei . T . evansi is transmitted mechanically by biting insects or , in South America , alternatively by vampire bats [26] . T . equiperdum is transmitted sexually during intercourse in horses [9] . Tsetse-independent transmission enabled these parasites to move out of the tsetse fly belt in sub-Saharan Africa . Mechanical transmission is a non-specific process that can take place when a vector undergoes interrupted feeding between hosts . Although any biting insect could transmit T . evansi from one host to the next , the insects responsible for most of its transmissions are haematophagous insects , such as horseflies and stable flies [32] . In addition to their ability to bypass the tsetse fly vector , all T . evansi ( and T . equiperdum ) strains analyzed so far are also characterized by having no or dysfunctional kinetoplast DNA , a trait referred to as dyskinetoplastidy [10 , 16 , 18 , 33] . Where present , kDNA has suffered homogenization of the minicircle component , which consists of more than 200 distinct classes in a tsetse transmission competent strain of T . brucei [34] . In all T . evansi strains analyzed to date , kDNA is dominated by either type A or type B minicircles [10 , 18 , 22 , 35] . Minicircle heterogeneity is essential for mitochondrial gene expression in trypanosomes [25] . As a consequence of its dyskinetoplastidy , T . evansi can therefore no longer complete cyclical development in the tsetse fly , and this could be one of the driving forces for the switch to mechanical transmission [16 , 18 , 36] . Another consequence of their inability to complete their development in tsetse flies is that both T . evansi and T . equiperdum strains do not undergo sexual reproduction . Although these peculiarities unite all T . evansi ( and T . equiperdum ) strains , there is significant variation in other traits such as virulence among parasite strains and animal host species [37] . In this study , we screened for genetic variation at a set of 15 highly variable polymorphic loci in a group of 35 T . evansi isolates from Kenya ( Fig 1 , Table 1 ) . In this area both T . evansi and T . brucei co-occur , making it a potential area where the trypanosome host shift into camels might have occurred [38] . The climate of this region is semi-arid and supports husbandry of both camels , the typical host of T . evansi in this region , and cattle and goats [2] , common hosts of T . brucei . The goal of this paper is to quantify levels and patterns of inter-strains genetic diversity among to understand the evolutionary origin of different T . evansi strains . This will help control and monitor disease spread by providing data that inform on the rate and modality of novel genotypic combinations that exists in the circulating T . evansi strains . Furthermore , this data provides general insights on the different ways T . brucei strains can evolve into epidemiologically novel parasites despite their very similar genetic background . This general phenomenon has important epidemiological implications for both the animal and human diseases that they cause .
For the purpose of this work , and in line with microbiological convention , we have defined the terms isolates and strains as follows . An isolate was obtained by sampling a particular animal at a particular point in time . A strain is an isolate or group of isolates that can be distinguished from other isolates by phenotypic and or genotypic characterization [39] . We analyzed a total of 41 T . evansi isolates . The majority of these isolates are from Kenya ( Fig 1 ) and currently stored at the KETRI cryobank [40] at KALRO-BRI ( Kikuyu , Kenya ) . These samples had been collected at several time points and some had previously been classified as T . evansi based on host species ( camel vs . non-camel ) , region of isolation , and kDNA minicircle type ( Table 1 ) . The virulence of two of these isolates , K2479 [19 , 41] and K3576 , were experimentally characterized in mice , based on relative levels of parasitemia and host survivorship in infected mice ( Kamidi et al . , in prep ) . The remaining T . evansi isolates came from multiple sources ( Table 1 , S1 Table ) and have been well-characterized in past studies and , in some cases , were part of recent genetic studies [10 , 15 , 18 , 19 , 36] . To provide a spatial breadth to our study and to be able to connect it with previous microsatellite analyses we also included 66 T . b . brucei and T . b . rhodesiense isolates ( S1 Table ) from across sub-Saharan Africa that have also been extensively characterized [10 , 15 , 42] . These isolates included at least one representative from each of the genetic clusters previously identified in sub-Saharan Africa [15] . Thus , the final sample set consisted of 107 T . brucei and T . evansi isolates , including 4 from buffalo in Asia and 103 from a variety of mammalian hosts in Africa , with a special focus on isolates from camels ( Fig 1 , Table 1 ) and wildlife ( S1 Table ) in Kenya . DNA was extracted from isolates that did not have DNA available using either the Qiagen DNeasy Blood and Tissue Kit ( Qiagen , Germany ) , following manufacturer’s protocols , or a phenol and chloroform protocol for samples for which DNA extractions were already available [43] . To further classify presumptive T . evansi samples not previously well classified [18 , 19 , 36]; we carried out a set of four diagnostic PCR tests for 37 isolates including 34 isolates for which we did not have certain classification ( Table 1 ) . First , we used PCR amplification of a 480 bp fragment of the Internal Transcribed Spacer ( ITS1 ) of the ribosomal DNA[44] , to confirm all isolates were pathogenic African trypanosomes . We then used PCR amplification of a 284 bp fragment of the serum resistance-associated ( SRA ) gene [45] , to confirm isolates were not T . b . rhodesiense . Then , we performed a PCR assay to identify isolates with the VSG antigen type RoTat 1 . 2 , used in serological and PCR-based diagnosis , that targets a 488 bp fragment of the RoTat 1 . 2 variant , as per previous protocol [23] . Although this gene occurs in most T . evansi type A [23 , 46] , it has been reported that T . evansi type B and some T . evansi type A strains may not have it [46 , 47] . In addition , T . evansi strains can lose the kinetoplast entirely [10 , 16] which would lead to a false negative result in a diagnostic PCR assay for type A minicircles . Thus , as an alternative to identify type A T . evansi , we designed a novel PCR assay . This assay targets a 3-bp deletion ( GTC codon , corresponding to alanine 281 ) in the nuclear encoded subunit γ ( systematic TriTrypDB ID Tb927 . 10 . 180 ) of the mitochondrial FOF1-ATPase . This deletion is unique to all T . evansi type A screened so far and to some closely related strains that had been classified as T . equiperdum [10] . This mutation is critical to compensate for loss of functional kinetoplast DNA in this group of T . evansi/T . equiperdum [48] . The assay consists of two PCR reactions , a diagnostic and a control PCR reaction ( S1 Fig ) . The diagnostic reaction ( using primer combination F1/R1 ) is designed to amplify an 855 bp fragment of FOF1-ATPase subunit γ , if at least one allele in the strain has this 3-bp deletion ( named A281del ) . The control PCR reaction ( using primer combination F1/R2 ) amplifies an 863 bp long fragment of the same region , regardless of kDNA type . Both PCR reactions were carried out in 10 μl volumes consisting of 5 μl 2X Type-It ( Qiagen ) , 0 . 25 μM of each primer , 10 ng of genomic DNA and dH2O . A touchdown thermal cycling protocol included a 5 min initial denaturation at 95°C , 10 cycles touchdown ( 95°C for 30 sec , 50°C minus 1°C per cycle for 30 sec , and 72°C for 1 min ) , and 30 cycles amplification ( 95°C for 30 sec , 40°C for 30 sec , and 72°C for 1 min ) , followed by a 7 min final extension period . All PCR runs included the isolates RoTat1 . 2 ( OB106 ) , a T . evansi type A , and cp24 , a T . b . brucei from Balmer et al [15] , as positive and negative controls , respectively ( Table 1 ) . We used fifteen microsatellite loci extensively validated in previous studies and using the same previously published protocols[49 , 50] . Primer sequences for amplification and chromosomal locations of the loci can be found in S2 Table . Amplifications were performed with fluorescently labeled forward primers ( 6-FAM and HEX ) using a standard PCR in 13 μl reaction volumes containing approximately 100 ng of genomic DNA , 5 μl of Type-it Master Mix ( Qiagen , Germany ) and 1 μl each of forward and reverse primers ( 10 μM starting concentration ) . PCR products were then multiplexed , combined with size standard ( Applied Biosystems ROX500 ) and highly deionized formamide , and genotyped on an ABI 3730xl DNA Analyzer ( Applied Biosystems Inc , USA ) at the DNA Analysis Facility on Science Hill at Yale University ( http://dna-analysis . yale . edu/ ) ) . Alleles were scored using the program GeneMarker v 2 . 4 . 0 ( Soft Genetics , State College , PA , USA ) with manual editing of the automatically scored peaks . To evaluate evolutionarily distinct genetic clusters within our dataset , we included all 107 T . b . brucei , T . b . rhodesiense , and T . evansi isolates in Bayesian cluster analyses using STRUCTURE v2 . 3 . 4 [51] . STRUCTURE runs indicated a K value ( number of clusters ) of less than ten . Thus , we performed 20 runs with a burn-in of 5 , 000 and a total of 250 , 000 iterations to assess the optimal K value with the Evanno method [52] , using the Clustering Markov Packager Across K ( CLUMPAK ) [53] . For final assignments of isolates to clusters , we performed 10 runs for K values one through ten with a burn in of 50 , 000 and 250 , 000 iterations . Each isolate was assessed for probability of assignment ( Q ) to each of the K clusters identified in the STRUCTURE analysis . We considered Q>0 . 80 as a “certain assignment” , and Q<0 . 80 as an “uncertain assignment" . We further evaluated evolutionary relationships and the levels of genetic differentiation among and within T . evansi and T . brucei genetic clusters and isolates of uncertain assignment using principal components analysis ( PCA ) of microsatellite data in the “adegenet” package in R v3 . 0 . 2 ( R Development Core Team ) . We estimated the centroid and region encompassing 95% of the variance observed within T . brucei subgroups identified in the STRUCTURE analysis . In order to compare levels of genetic diversity and differentiation among T . evansi isolates with those found among T . brucei ( T . b . brucei + T . b . rhodesiense ) isolates , we estimated levels of diversity within the STRUCTURE defined clusters as well as levels of differentiation between and within clusters . For these analyses , we included only isolates with high probability of assignment ( Q > 0 . 80 ) to STRUCTURE-based clusters at three levels: ( i ) all isolates regardless of taxonomy , ( ii ) T . brucei isolates only , and ( iii ) T . evansi isolates only . To understand diversity within clusters at these three levels , we estimated allelic richness ( AR ) in FSTAT v1 . 2 [54] , observed and expected heterozygosity ( HO and HE ) and the related Fisher’s inbreeding coefficient ( FIS ) in the R package HIERFSTAT v0 . 4–10 [55] . To understand patterns of within-cluster genetic distance at these three levels , we calculated pairwise genetic distance between isolates using the Reynolds distances [56] . We estimated a distance tree using the UPGMA method implemented in the “PopPR” v2 . 3 . 0 package [57 , 58] in R with 1000 bootstrap replicates . We then tested for significant differences in within-cluster genetic distances with an analysis of variance ( ANOVA ) followed by a Tukey-Kramer HSD test performed in JMP v11 . 2 ( SAS Institute Inc . , Cary , NC , USA , 1989–2012 ) . To ensure that the time of isolation did not account for cluster assignment , we used the software JMP to perform a Chi-square test of the time of isolation ( by decade ) , with the taxon of each sample included as a co-variate . Finally , to understand patterns of among-cluster differentiation at the same three levels , we estimated pairwise FST in ARLEQUIN v . 3 . 5 [59] with Wright’s statistics [60] , following the variance method [61] , using 10 , 000 permutations , 1 , 000 , 000 Markov chain steps , and 10 , 000 dememorization steps to obtain exact p-values .
Results from the PCR assays are presented in Table 1 . We found that all of the KETRI isolates amplified in the PCR test that is diagnostic for the ITS1 region of all African trypanosomes considered pathogenic: Members of the subgenera Nannomonas ( T . congolense ) , Duttonella ( T . vivax ) and Trypanozoon ( T . brucei , T . evansi , T . equiperdum ) [22] . In contrast , T . lewisi and T . theileri , which are considered non-pathogenic but can be found in many areas of the world , including Kenya , have been reported to not give a positive signal , presumably because their ITS region is more divergent [22] . All isolates were also SRA negative , confirming the absence of T . b . rhodesiense isolates . For the A281del . PCR assay , five isolates could not be determined because they failed to amplify in the positive control reaction ( n/a in Table 1 ) . Of those that amplified , we found 29 isolates to be A281del positive , indicating that they are T . evansi type A , and 3 isolates that were A281del negative , indicating that they could be either type B or something else , but not type A . Only 20 of the isolates tested were positive for the RoTat 1 . 2 gene ( including , as expected , STIB810 and C13 ) , indicating a diversity of VSG antigen types in our dataset . Although it has been reported that type A T . evansi isolates are typically RoTat1 . 2 positive [10 , 21 , 24] , we found that of the 29 A281del positive isolates , only 17 were RoTat 1 . 2 positive while 12 were RoTat 1 . 2 negative ( Table 1 ) . The combination of these PCR assays suggests that , at least in Kenya , T . evansi isolates that are type A but RoTat1 . 2 negative are more prevalent than expected [23 , 46 , 47] , which could result in a considerable frequency of false negatives for current diagnostic tools for surra [23 , 24] . The results suggested a K-value of 2 , and thus the presence of two distinct genetic clusters , as the most likely hierarchical level of population structure that best fits the method’s assumptions ( S2 Fig ) . One of these two clusters ( S3 Fig; top panel , orange color ) includes most but not all T . evansi isolates , while the other includes all of the T . brucei brucei and T . b . rhodesiense isolates ( S3 Fig; top panel , blue color ) . The next best fit of K = 7 was able to distinguish structure within T . brucei , suggesting the presence of seven distinct genetic units . Assignment to these clusters for the 107 isolates analyzed is shown in Fig 2A and S3 Table . While the majority of the isolates ( 78% ) had a high level of assignment to only one cluster ( Q > 0 . 80; colors in bars in Fig 2 represent scores listed in S3 Table ) , 7 T . b . rhodesiense , 13 T . b . brucei , and 3 T . evansi isolates showed uncertain assignment to any one of seven clusters ( Q < 0 . 80 , bars with no single color representing more than 80% in Fig 2 ) to any one of seven clusters ( Fig 2A ) . This uncertain assignment could be due to a variety of factors , ranging from shared common ancestry or recent admixture to limitations of the genetic markers to separate such recently diverged taxa . Cluster “b” ( purple ) includes only T . b . brucei isolates and corresponds to the “Kiboko B” group [15] . Cluster “a” ( orange ) , “c” ( blue ) , “d” ( green ) , and “f” ( grey ) include both T . b . brucei and T . b . rhodesiense isolates . Cluster “g” ( red ) includes isolates from all the three taxa , T . b . brucei , T . b . rhodesiense , and T . evansi . Cluster “e” ( yellow ) includes only T . evansi isolates . The level of population structure and grouping we observed for T . brucei is similar to results from previous microsatellite [15 , 42] and genomic [13 , 14] analyses , where T . b . rhodesiense isolates were consistently assigned to multiple clusters together with T . b . brucei isolates . This data confirms multiple independent origins of the human disease parasite , T . b . rhodesiense , from different non-human infective T . b . brucei strains and implies that the SRA gene has moved horizontally between strains , which is consistent with earlier studies and experimental evidence that this can occur in the field [6 , 13 , 14 , 15 , 42 , 62 , 63 , 64 , 65 , 66] . As pointed out previously , this finding has important practical implications for disease control and monitoring , as it provides further evidence that T . b . brucei strains can relatively easily transform into T . b . rhodesiense strains and pose a serious risk to human health [13 , 14] . The STRUCTURE results for T . evansi isolates are displayed in detail in Fig 2B . Also included are the results of the RoTat 1 . 2 PCR assay and information on the kDNA minicircle type ( based on the literature , where available , or as predicted from our A281del PCR assays; see Tables 1 and S1 ) . Although the majority of T . evansi isolates assigned to cluster “e” ( yellow ) , there are 6 isolates that assigned with high Q values ( Q > 0 . 80 ) to different STRUCTURE-defined genetic clusters , and 3 isolates ( STIB810 , STIB708 and STIB806K ) with uncertain assignment ( Q < 0 . 80 ) . Of the isolates with high Q values to non “e” clusters , one isolate ( K2479 ) assigned to cluster “f” ( gray ) , two isolates ( K3552 and K3557 ) to cluster “c” ( blue ) , and two isolates ( RoTat1 . 2 and STIB811 ) to cluster “g” ( red ) , implying that some T . evansi isolates are genetically closer to T . brucei isolates than to each other and supporting the hypothesis of multiple independent origins of T . evansi isolates from T . brucei . All 33 isolates with kDNA minicircle type A were assigned to either cluster “e” or “g” , the single confirmed type B ( K2479 ) assigned to cluster “f” , and the two isolates that could not be classified as type A or type B by our assays ( K3552 and K3557 ) assigned to cluster “c” ( Fig 2B ) . This result suggests an association of kDNA minicircle type A with the “e” and “g” clusters , and that the other isolates in our dataset associated with other dominant minicircle types ( S1 Table ) are from genetically distinct lineages . In contrast , there was no assignment pattern for the isolates that typed as RoTat 1 . 2 positive or negative based on the PCR assay ( Table 1 ) , as the positive isolates assigned to three different clusters ( “c” , “g” , and “e”; Fig 2B ) . The high virulence isolate , K2479 ( a kDNA minicircle type B and RoTat 1 . 2 negative isolate ) , grouped with the “f” cluster , while the low virulence isolate , K3576 ( a RoTat 1 . 2 positive isolate ) assigns to the “e” cluster ( Fig 2B ) . This separation into different clusters suggests independent evolution , but more samples from different genetic backgrounds and virulence degrees are necessary to validate the generality of this observation . The results of the multivariate analyses ( PCA , Fig 3 ) largely confirmed the pattern of genetic structuring suggested by the Bayesian analyses ( Fig 2A and 2B ) and also provided additional insights on how the different STRUCTURE-based clusters are genetically similar . Individuals from four of five STRUCTURE-defined clusters that include both T . b . brucei and T . b . rhodesiense isolates ( clusters “a” , “c” , “d” , “f” , and “g” ) grouped close together in the multivariate space defined by the first two PC axes , with isolates from the “a” and “g” , and isolates from the “c” , “d” , and “f” clusters being indistinguishable from one another along the first two components ( PC 1 and 2 ) . These close genetic relationships were also implied by the uncertain STRUCTURE cluster assignment of some T . brucei , which suggests some shared ancestry with all these clusters ( Fig 2A , bars with no single dominant color representing more than 80% of the size ) . On the other hand , the T . b . brucei “Kiboko B” isolates ( cluster “b” , Fig 2A ) were clearly genetically distinct from the other isolates ( purple ellipsoid in Fig 3 ) , as also suggested by the high Q values assignment of these isolates to a single STRUCTURE-based cluster ( Fig 2A ) . The isolates included in STRUCTURE-based cluster “e” ( yellow in Fig 2A , exclusively T . evansi isolates ) , were also separate from the others . However , they were proximal to cluster “g” isolates and to two T . evansi isolates with uncertain assignment ( Fig 2B ) , indicating a close evolutionary relationship between the T . evansi and T . brucei isolates in these two clusters ( Fig 3 ) . As for the STRUCTURE analyses , some T . evansi isolates were closer to T . brucei isolates included in different clusters ( “f” , “c” , and “g"; Fig 2B ) . Thus , both Bayesian and multivariate analyses suggest that some T . evansi isolates share closer evolutionary relationships with different T . brucei isolates than with each other . To compare diversity and differentiation within and among T . evansi and T . brucei , we estimated basic diversity statistics , genetic distance , and FST among STRUCTURE-based clusters at three levels defined as follows: ( i ) all of the 84 isolates with Q > 0 . 80 regardless of taxonomy ( S3 Table ) , ( ii ) the 46 T . brucei isolates with Q > 0 . 80 ( S3A Table ) , and ( iii ) the 38 T . evansi isolates with Q > 0 . 80 ( S3B Table ) . Basic diversity statistics are shown in Table 2 . Allelic richness within clusters of all isolates ( Table 2A ) ranged from 2 . 10 in cluster "d" to 3 . 86 in cluster "f" , indicating the lowest genetic diversity in cluster “d” that contains both T . b . brucei and T . b . rhodesiense , but not T . evansi ( Fig 2 ) , and the highest genetic diversity in cluster “f” that contains T . b . brucei , T . b . rhodesiense , and T . evansi ( Fig 2 ) . Observed and expected heterozygosity levels and the related inbreeding coefficient ( FIS ) are also reported in Table 2A . Within clusters including all isolates , HO ranged from 0 . 50 in cluster “g” to 0 . 66 in cluster “e” , HE ranged from 0 . 47 in cluster “d” to 0 . 78 in cluster “g” , and FIS ranged from -0 . 30 in cluster "e" to 0 . 34 in cluster "f" , spanning a wide range of heterozygosity and conformity to the expectations of Hardy-Weinberg ( H-W ) equilibrium . This is not surprising given the importance of random mating and sexual reproduction in the maintenance of H-W equilibrium , and the known variation of these life history traits among trypanosome taxa [67 , 68] . For T . brucei only isolates ( Table 2B ) , within cluster allelic richness estimates were very similar but slightly lower than the estimates based on all isolates ( Table 2A ) . HO ranged from 0 . 50 to 0 . 63 , HE ranged from 0 . 47 to 0 . 76 , and FIS values were mostly positive , ranging from -0 . 18 to 0 . 36 ( Table 2B ) . Thus , T . brucei observed and expected hetrozygosity and FIS values indicate moderate deviation from H-W expectations , and are similar to those reported in a previous study [42] , where FIS ranged from -0 . 16 to 0 . 43 . For T . evansi only isolates ( Table 2C ) , within cluster allelic richness was intermediate to that found in T . brucei , indicating genetic diversity similar to that found in T . brucei . HO ranged from 0 . 40 to 0 . 69 , HE ranged from 0 . 36 to 0 . 72 , and FIS values ranged from -0 . 30 to 0 . 19 . Negative FIS in some clusters in both T . brucei and T . evansi could result from clonal , non-sexual reproduction ( as expected for the latter ) because there is a well understood decrease in expected heterozygosity during clonal reproduction , which lowers FIS [69] . The finding of relatively high allelic richness in all clusters and both positive and negative FIS values in both T . brucei ( Table 2B ) and T . evansi ( Table 2C ) could be a reflection of different relative levels of sexual and clonal reproduction and recombination among T . brucei isolates in different clusters , and to the fact that for T . evansi isolates are strictly clonal . To evaluate if levels of genetic differentiation among T . evansi isolates were different from the ones observed among T . brucei isolates , we estimated pairwise genetic distances , using Reynolds distances . First , we estimated a distance tree using all the 107 isolates ( S4 Fig ) . This tree clustered the T . evansi isolates in four different groups , confirming the results of both Bayesian and multivariate analyses ( Figs 2 and 3 ) , although bootstrap values among these groups are not high , thus limiting the strength of the inference that can be drawn from this analysis . Next , we estimated within-cluster distances using the STRUCTURE-defined clusters , including only the 84 isolates with Q > 0 . 8 ( S3 Table ) , as described for the estimates of basic diversity statistics ( Table 2 ) . Within-cluster mean distances among all isolates ( S4A Table ) averaged 0 . 70 and ranged from 0 . 57 in cluster “e” to 0 . 80 in cluster “f” , indicating that the lowest within-cluster distance occurs in the T . evansi only cluster , and the highest within-cluster distance occurs in a cluster that contains T . b . brucei , T . b . rhodesiense and T . evansi of type B . Within-cluster mean distances among T . brucei isolates averaged 0 . 72 and ranged from 0 . 61 in cluster “d” to 0 . 81 in cluster “f” ( S4B Table , S5 Fig ) . Finally , within-cluster mean distances among T . evansi isolates averaged 0 . 64 and ranged from 0 . 57 in cluster “e” to 0 . 75 in cluster “g” ( S4C Table , S5 Fig ) . The implications of these findings for evolutionary origins of T . evansi are discussed in detail below . The analysis of variance ( ANOVA ) indicated that within-cluster distance was significantly dependent on cluster of assignment ( p-value < 0 . 0001 ) . The results of the Tukey-Kramer HSD test are reported in S5 Table . These tests indicated that T . evansi cluster “e” and T . brucei cluster “d” had significantly lower within-cluster distance than any other cluster ( S4 Table , S5 Fig ) , suggesting that the most common T . evansi lineage ( cluster “e” ) is of recent origin and is made up of more closely related isolates than those included in most T . brucei clusters ( except cluster “d” ) . However , since this test could only be carried out for one of the T . evansi clusters , cluster “e” , because of low number of T . evansi isolates in the other clusters , the generalitiy of this finding remains uncertain without further sampling of a greater diversity of T . evansi isolates from non “e” clusters . To compare among-cluster differentiation in T . evansi and T . brucei , we estimated among-cluster FST using the STRUCTURE-defined clusters and only including the 84 isolates with Q > 0 . 8 ( S3 Table ) , as described for the estimates of basic diversity statistics ( Table 2 ) . FST estimates are reported in S6 Table . Among-cluster FST estimates between clusters regardless of taxonomy ( S6A Table ) ranged from 0 . 08 between clusters “g” and “e” to 0 . 31 between clusters “a” and “d” and showed significant differentiation between all clusters ( p-value < 0 . 006 ) , indicating that the lowest genetic differentiation was found between two clusters that contained T . evansi ( “g” and “e” ) , and that the highest genetic differentiation was found between two clusters ( “a” and “d” ) made up of entirely T . brucei isolates . Thus , the most common T . evansi cluster “e” is less differentated from the T . brucei-only cluster “a” than both T . brucei-only clusters “a” and “d” are to one another . Among-cluster FST estimates in T . brucei ( S6B Table ) ranged from 0 . 10 to 0 . 31 ( S6B Table ) , and showed significant differentiation between all clusters ( p-value < 0 . 005 ) , indicating high levels of genetic differentiation . Among-cluster FST in T . evansi ( S6C Table ) were similar to those in T . brucei , ranging from 0 . 06 to 0 . 29 , and showed significant differentiation ( p-value < 0 . 0001 ) between T . evansi in all clusters except the least differentiated clusters “e” and “g” , suggesting T . evansi cluster “e” and “g” are not significantly differentiated from each other . The low sample size of T . evansi in cluster “g” remains another possible reason for the non-significant p-value in FST estimates between “e” and “g” , and again highlight the need for further sampling of a greater diversity of T . evansi strains from non “e” clusters . These results indicate that the genetic diversity across all T . evansi isolates ( “overall” in Tables 2C and S4C ) represents a large amount of the genetic diversity found across T . brucei isolates ( “overall” in Tables 2B and S4B ) . However , within clusters including all isolates , the most common T . evansi cluster , cluster “e” , shows the least amount of genetic differentiation among isolates and the lowest amount of within-cluster genetic diversity compared to other clusters ( Tables 2A and S4A ) , with only the T . brucei cluster “d” showing similarly low levels ( Tables 2A and S4A ) . The Chi-square test showed that the time of isolation did not account for cluster assignment ( Chi2 = 20 . 19 , degrees of freedom = 30 , p-value = 0 . 9113 ) . Clustering and diversity analysis indicate that T . evansi strains likely originated from multiple genetic backgrounds ( Figs 2 and 3 ) and that the genetic diversity harbored by the T . evansi isolates analyzed in this study encompass a large proportion of the total diversity found in the T . brucei isolates ( Tables 2 and S4 ) . The single type B and the two unclassified isolates fall into distinct clusters ( "f" and "c" , respectively; Fig 2 ) , while type A isolates separate into two clusters ( "e" and "g"; Fig 2 ) , that are closely associated in the multivariate analysis ( yellow and red; Fig 3 ) . Cluster "e" is made up entirely of T . evansi isolates ( Figs 2 and 3 ) , while cluster "g" includes a mix of T . b . brucei , T . b . rhodesiense , and T . evansi ( Figs 2 and 3 ) . Separation of type A into two closely related clusters suggests that the T . evansi only cluster "e" has evolved from within cluster "g" , and both have evolved from the same T . brucei ancestor . Nonetheless , these results could also indicate that traits that are common between T . evansi in clusters "e" and "g" have evolved twice , independently . Evidence for these alternative hypotheses remains inconclusive . Support for a single origin of type A from within cluster "g" comes from the non-significant differentiation ( FST ) found between the T . evansi isolates in clusters "e" and "g" ( Fst = 0 . 06 , p-value = 0 . 105; S6C Table ) , which indicates high similarity between these clusters . Furthermore , certain T . evansi isolates from China ( STIB810 , STIB811 , and STIB806K ) that were isolated within 3 years from each other and presumably are closely related [10 , 17 , 36] can be found in both clusters "e" and "g": STIB810 assigns to cluster "e" , STIB811 assigns to cluster "g" , and STIB806K assigns about equally to both "e" and "g" ( Fig 2 ) , suggesting the “e” and “g” clusters are not the result of distinct geographic origins or outbreaks . Thus , distinct clustering of type B in cluster "f" , distinct clustering of unclassified isolates in cluster "c" , and nested clustering of type A isolates in the two closely related clusters "e" and "g" suggests independent origins of each T . evansi kDNA type from a diverse T . brucei background . The results from our screen of 15 microsatellite loci largely aligns with previous phylogenetic and population genetic analyses , which indicated that T . evansi strains are nested phylogenetically within the more genetically diverse T . brucei [1 , 10 , 11 , 13 , 17 , 70] , likely originated from different T . b . brucei strains [10 , 70] , and are highly variable [35 , 70] . Some studies [70–72] found that the T . evansi strains sampled clustered closely with one another and separately from T . b . brucei and T . b . rhodesiense strains . We suggest that this pattern of genetic similarity can be an artifact resulting from the limited number and type of isolates included in these studies . This is especially true for the T . evansi isolates that only included the common kDNA type A lineage ( i . e . kDNA minicircle type A configuration and RoTat 1 . 2 positive ) . Indeed , other studies that have included both type A and type B T . evansi isolates have found similar results to what we have found , using a larger geographic and taxonomic diversity of isolates [10 , 17 , 35 , 73] . Interestingly , our findings are also consistent with previous comparative genomic analysis [10] and with classical parasitological characterization , which indicates high similarity between T . evansi and T . b . brucei except for variable patterns of loss of part or all of their kDNA [1 , 12 , 16 , 74] .
This work shows that T . evansi strains from Eastern Africa , the main region where both T . evansi and T . b . brucei strains co-occur , likely originated from multiple T . b . brucei strains and harbor a high degree of circulating genetic variation . This result is surprising because of the phenotypic similarities between all T . evansi strains , such as ability to sustained mechanical transmission outside the tsetse belt , variable loss of functional kDNA , and the common disease symptoms they cause in a variety of animals . Multiple origins of T . evansi phenotypes implies that complex traits such as ability for mechanical transmission have evolved multiple times , and that there is plenty of standing genetic diversity to provide opportunity for selection to generate novel strains . Further research is needed to understand the mechanism of this evolutionary transition . Our results provide further support for the idea that the taxonomic rank of T . evansi is not valid from an evolutionary standpoint [10 , 12 , 17 , 75] . However , even the subspecies designation suggested by some authors is not taxonomically correct , since this rank should , by definition , be used to identify groups of populations within a species that are geographically and genetically differentiated . We propose that the taxonomy of the groups within the genus Trypanosoma , including T . b . rhodesiense , T . evansi and T . equiperdum , requires a fundamental revision that , as proposed by Gibson [67] , should ‘bring together considerations of utility , genetic difference and adaptation’ . These findings mirror what is known about the multiple evolutionary origins of T . b . rhodesiense from different strains of the animal parasite T . brucei brucei , and thus highlight the trypanosome’s ability to evolve novel and complex traits to expand their host repertoire . This has important epidemiological implications , as T . b . brucei strains from different genetic backgrounds apparently can become either parasites of a lethal human disease ( i . e . T . b . rhodesiense ) [76 , 77]or become able to be transmitted by a variety of hematophagous insects besides the tsetse fly ( i . e . T . evansi ) [10 , 68 , 76] . To date , there have been only few reported cases of T . evansi infecting humans [78] a well-documented case from India was thought to be non-transmissible to other humans with fully functioning immune systems [79] . Thus , risk of human infective T . evansi remains theoretical , but deserves consideration since this would allow human sleeping sickness to escape sub-Saharan Africa and take advantage of hosts worldwide . In order for the human disease to escape sub-Saharan Africa , both mechanical transmission and evasion of the human immune system would be needed in a single strain . The fact that trypanosomes have been able to acquire both traits repeatedly makes the acquisition of both features in one strain a dangerous possibility . However , this possibility remains remote for several reasons . First , mechanical transmission in human infective strains would require much higher levels of parasitemia than observed in infections caused by T . b . gambiense [7 , 80] , the subspecies responsible for the vast majority of cases of human African trypanosomiasis . Second , the acquisition of the SRA gene requires sexual recombination in the tsetse fly , which does not occur in T . evansi once it has become dyskinetoplastic . Nonetheless , if this were to happen , the spread of sleeping sickness outside of sub-Saharan Africa would have dramatic consequences because diagnosis is complicated , pharmacological therapy is inadequate [81–83] , and vaccines are non-existent . Future work should therefore focus on understanding the origin and dynamics of the T . evansi spatial expansion from Africa to multiple continents , as well as on the functional and molecular basis of the ability to by-pass tsetse flies for their transmission . Screening for genetic polymorphism in additional T . evansi isolates from across the world will help us understand the origin and timing of the T . evansi expansion , evaluate if only a few genetically similar strains were responsible for the spread , and identify the T . brucei genetic background most likely to give rise to T . evansi strains . Adding genome-wide data will provide higher resolution of the phylogenetic relationships among these strains and insights on the genetic , functional and molecular basis of novel complex traits such as “mechanical transmission” . | Trypanosoma evansi is an important pathogen of the camel and other livestock where it is a causative agent of surra ( an economically burdensome disease ) . The T . evansi is found in Kenya and the rest of the world . This study indicates that T . evansi originated recently from multiple Trypanosoma brucei strains from different genetic backgrounds . This suggests multiple independent evolutionary origins of some complex traits that may have facilitated mechanical transmission in T . evansi and subsequently enabled the parasite to escape the obligate link with the African tsetse fly vector . This evolutionary origin appears to have occurred repeatedly . Our results provide a more comprehensive understanding of the epidemiology of surra , provide recommendations for future work , and indicate a need to consider the risk of horizontal transfer of epidemiologically relevant traits among different Trypanosoma genetic backgrounds in any control campaign . Thus , our study is an important contribution to the field , and represents an important step towards the ultimate aim of trypanosomiasis prevention and/or elimination . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"and",
"discussion",
"Conclusions",
"and",
"future",
"directions"
] | [
"parasite",
"evolution",
"vertebrates",
"parasitic",
"protozoans",
"mammals",
"animals",
"parasitology",
"organisms",
"trypanosoma",
"brucei",
"protozoans",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"kinetoplasts",
"camels",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"evolutionary",
"genetics",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"trypanosoma",
"brucei",
"gambiense",
"evolutionary",
"biology",
"amniotes",
"polymerase",
"chain",
"reaction"
] | 2017 | Multiple evolutionary origins of Trypanosoma evansi in Kenya |
Vibrations through substrates are an important source of information for diverse organisms , from nematodes to elephants . The fundamental challenge for small animals using vibrational communication is to move their limited mass fast enough to provide sufficient kinetic energy for effective information transfer through the substrate whilst optimising energy efficiency over repeated cycles . Here , we describe a vibratory organ found across a commercially important group of plant-feeding insects , the planthoppers ( Hemiptera: Fulgoromorpha ) . This elastic recoil snapping organ generates substrate-borne broadband vibrations using fast , cyclical abdominal motion that transfers kinetic energy to the substrate through the legs . Elastic potential energy is stored and released twice using two different latched energy-storage mechanisms , each utilising a different form of elastic recoil to increase the speed of motion . Comparison to the acoustic tymbal organ of cicadas ( Hemiptera: Cicadomorpha ) reveals functional convergence in their use of elastic mechanisms to increase the efficacy of mechanical communication .
We begin by characterising the morphology of the newly-described snapping organ in our model species , Agalmatium bilobum ( Fulgoromorpha: Issidae ) . The snapping organ can be found dorsally on each side of the body at the junction between the metathorax and the abdomen , spanning the first two abdominal segments ( Fig 1A and 1B and S1 Movie ) . The organ has a W shape; a ridge ( Fig 1B ) articulates at its base with the thorax ( first ‘V’ ) and fuses at its tip to the anterior arm of a Y-lobe ( Fig 1B ) , which has resilin ( Fig 1B ) between its arms ( second ‘V’; Figs 1B and S1 ) . The posterior arm of the Y-lobe is fused with the second segment , tergum 2 ( tg2; Fig 1B ) of the abdomen . The Y-lobe is linked at its base to an internal spine ( sp; Fig 2C ) of the second segment via a membranous connector ( Figs 1B , 2A and 2C ) . Eight muscle pairs are directly associated with the snapping organ ( Fig 2 and S2 Table and S1 Text ) , comprising three pairs of dorsal longitudinal muscles ( DLMs ) and five pairs of dorsoventral muscles ( DVMs ) . Four other muscle pairs are indirectly associated with the snapping organ ( ventral longitudinal muscles [vlms] IIIvlm2 , Ivlm1 , and IIvlm2 and intersegmental dorsoventral muscle [IIisdvm] ) ( Fig 2A ) . The snapping organ is not sexually dimorphic . Homologous vibrational organs are present throughout the entire planthopper clade ( Fig 3 and S1 Table ) . The defining features of the musculature ( Fig 2 and S2 and S3 Tables ) , innervation ( S2 Fig ) , and external morphology ( the ridge , Y-lobe , and connector ) of the snapping organ are consistent and identifiable , despite variation in its proportions and shape across the planthoppers ( Fig 3B–3E and S1 Text ) . Two groups deviate from this general picture: part of the family Delphacidae , in which the exoskeleton and musculature have been drastically reorganized to form an entirely different type of vibrational organ ( S3 Fig and S1 Text ) , and part of Derbidae , which have an externally obscure snapping organ and also possess tentative stridulatory structures ( Fig 3 and S4 Table ) . Based on their phylogenetic position , the deviations observed in these two groups are likely to be derived ( Fig 3 ) . To determine the kinematics of the snapping organ , we used high-speed videography and laser vibrometry on our model species , A . bilobum ( Fig 4 and S1 Movie and S1 Data ) . Each vibrational cycle began with the snapping organ in its relaxed position ( Figs 1D and 4A ) . Subsequently , the thorax/midabdomen was raised over a 15 ms timescale ( Fig 4B ) . The first mechanical impulse followed ( loading vibrational peak ) , which resulted in closed Y-lobe arms , extended ridge , and the base of the Y-lobe pulled down and rotated clockwise ( Fig 4C ) . The system resonated in response , giving a jagged waveform over a 15–20 ms timescale ( Fig 4D ) . The cycle was completed by a second mechanical impulse ( unloading vibrational peak ) , in which the Y-lobe arms reopened , the ridge retracted , and the base of the Y-lobe rose and rotated back ( Fig 4E ) . This resulted in whole-system resonance ultimately returning the organ to the same relaxed position as at the beginning of the cycle . Each vibration generation cycle takes place within 120 ms , and the mechanism does not generate any audible acoustic noise [27] . We propose that each cycle of vibration generation consists of four main steps ( Figs 1D and 4 ) . Transition from the relaxed state to the cocked state was comparatively slow ( on a timescale of 15 ms ) , and the movements of landmarks on the external exoskeleton suggest that this phase of the cycle was driven directly by DLM contraction ( Figs 2 and 4B ) . Whilst we do not have direct recordings of muscle activity , the distance between the origin and insertion points of both DLMs shortens at this point in the cycle ( Fig 4B ) , and there is no other muscle whose action could produce this strain . The distance between these points shortens even further at the transition from the cocked state to the loaded state ( Fig 4C ) , but this change occurs too quickly to be explained by direct muscle action alone . Specifically , the rate of change in the kinetic energy of the abdomen during loading implies energy release at a much higher power density than the DLMs and DVMs combined ( Idlm1 , Idlm2 , IIedvm1 , IIedvm2 ) could possibly supply through contraction ( 7 , 080 W kg−1 , which is nearly 15 times the typical 500 W kg−1 power density for a muscle [31]; see S1 Methods and S1 Data ) . It follows that some form of elastic recoil , which acts as a kind of mechanical power amplifier , must be involved in the transition between the cocked and loaded states . This fast phase ( 0 to peak velocity taking 0 . 35 ms ) , which we term loading , is responsible for producing the first mechanical impulse transferring vibrational energy to the substrate . The distance between the origin and insertion points of the DVMs also shortens at this point in the cycle ( Fig 4C ) , but contraction of these muscles alone cannot supply the mechanical energy at a high enough rate to explain the rapidity of the loading phase . Instead , the events at this transition are consistent with DVM contraction serving as an unlatching mechanism that triggers the rapid pulling down of the abdomen , followed by system resonance ( Fig 4C ) . The next phase of the cycle , in which the system transitioned to its pretrigger state , was a slow phase , probably involving muscle relaxation , over a 15–20 ms timescale . The subtle shift of exoskeleton positions , and particularly the lengthening of the distance between the points of origin and insertion of the DLMs ( Fig 4D ) , is consistent with the DLMs relaxing during this phase . In contrast , the distance between the points of origin and insertion of the DVMs remain constant through this phase of the cycle , suggesting that they remain in their contracted state . The final transition in the cycle was from the pretrigger state to the relaxed state . This second fast phase , which we term unloading , is responsible for producing the second mechanical impulse transferring vibrational energy to the substrate . The associated increase in distance between the points of origin and insertion of the DVMs ( Fig 4E ) suggests that unloading is triggered by DVM relaxation , which causes the rapid return of the snapping organ to its relaxed conformation through a second release of stored elastic potential energy . There is no evidence for muscle contraction at this phase of the cycle , and we therefore infer that this elastic potential energy is likely to be stored in the deformed exoskeletal elements of the snapping organ . To verify whether passive release of elastic potential energy could be responsible for the fast unloading phase , we built a simplified mathematical model of the snapping organ , in which we replaced the ridge and the anterior arm of the Y-lobe with a pair of rigid bars connected in series to the thorax by a pair of torsional springs ( Figs 5A and S4 ) . The stiffness constants of these torsional springs were determined experimentally in a static loading experiment ( S1 Methods ) . The abdomen and posterior arm of the Y-lobe were modelled as a mass-spring–damper system attached to the free end of the second rigid bar ( Figs 5A and S4 ) , and the spring constants and damping coefficients of this system were fitted as free parameters ( S1 Methods ) . Quantitative comparison of the measured and modelled motion supports our supposition that the unloading phase can be explained through passive recoil of the Y-lobe , in which mechanical energy is stored elastically ( Fig 5B and S1 Methods ) . When released , the elastic potential energy of these stiff springs acts to move the mass of the abdomen back to its relaxed state , resulting in resonant motion of the abdominal mass . More harmonic content is apparent in the measured vibrations than the modelled ones , which is not surprising given the simplicity of the model , but importantly from the perspective of information transfer , both the measured and the modelled spectra involve a broad range of different frequencies ( Fig 5C and 5D ) . The motion generated by the snapping organ during the two fast loading and unloading phases was on a timescale that would not have been possible through direct muscle action alone . The snapping organ instead uses two distinct elastic recoil mechanisms , each of which involves storing energy in springs , then releasing the stored energy quickly [8–11] . During the loading phase , the obvious candidate locations for elastic energy storage are the DLMs themselves , given that the exoskeleton itself deforms very little during loading ( Fig 4B ) . This would mean that these muscles act both as engines , actively generating the force required for loading , and as springs , storing elastic energy within their deformed structure when subject to resistance against shortening from the exoskeleton . Muscles have previously been suggested to act as springs [10] , and here the elastic energy storage is in the range achievable by the cross-bridges ( energy density for paired Idlm1 and Idlm2 conservatively c . 2 . 47 J kg−1 ) [32] . We therefore suggest that resistance to shortening of the contracted DLMs allows these muscles to act as an elastic spring during the loading phase [31] , storing energy slowly , then releasing this quickly when triggered . A latch must be involved to prevent early release of energy , and a mechanical constraint at the base of the Y-lobe could act as a latch that is removed when the DVMs contract , acting to trigger the release of elastic potential energy stored in the DLMs . During the unloading phase , a more straightforward passive elastic recoil is the likely mechanism , as captured by our mathematical model ( Fig 5 ) . Energy is stored in stiff springs within the W-shaped exoskeleton linkage system that are deformed and therefore loaded during the loading phase ( Fig 4C ) , but which return to their resting position and are therefore unloaded following the unloading phase ( Fig 4E ) . The first elastic recoil event during the active loading phase thereby stores the energy that is released during the second elastic recoil event , which is the passive unloading phase . DVM relaxation is the likely trigger , with the membranous connector and acting as a possible cuticular latch preventing early release ( Fig 2 ) . Rapid recoil is made possible by DLM relaxation during the pretrigger step , and resilin between the Y-lobe arms ( S1 Fig ) will act to limit damage during recoil . Additional muscles may modulate the vibration during unloading ( e . g . , IIIvlm2 ) , but the muscles are far too small to account for the power density during unloading ( c . 765 , 000 W kg−1 if normalising the mechanical power by IIIvlm2 mass; Fig 2A and S1 Data ) . In summary , the snapping organ uses two muscle contraction events per cycle and typically repeats its cycle every 0 . 3–1 s [33] , giving a muscle contraction frequency of under 5 Hz ( S5A Fig ) . In contrast , the frequencies of the mechanical impulses resulting from this motion as measured on the midabdomen were broadband under 3 kHz ( shown for recoil in Fig 5C and 5D ) . Crucially , from a communication perspective , the complete system also acts to transfer mechanical motion from the snapping organ to the substrate . This represents another form of mechanical power transformation , albeit one that is modulated by the substrate . For motion vertical to the plant stem for one individual , the velocity ratio of motion measured on the plant relative to motion measured on the insect midabdomen indicates that the velocity of motion is attenuated by 83% ( average −15 . 5 ± 6 . 2 dB ) , with lower attenuation in velocity of motion between the prothorax and plant at 71% attenuation ( average −10 . 5 ± 5 . 5 dB , S1 Data and S5 Fig ) .
The consistency of the snapping organ’s morphology , and its systematic distribution across planthoppers indicates that this most likely represents a conserved mechanism for generating abdominal vibrations across the Fulgoromorpha . Previous studies have only examined delphacid vibrational organs [24 , 26 , 34] , but our analysis of their peculiar morphology indicates that the drumming organs of delphacids are the exception and not the rule . The consistency of snapping organ morphology across the rest of the planthoppers provides a clear mechanistic explanation for the observed uniformity of their vibrational signals [25 , 33] . These findings reflect the fundamental importance of vibrational signals in planthopper communication . The functional morphology of the snapping organ also reveals some remarkable functional convergences and some equally remarkable mechanistic differences between the mechanical communication mechanisms of planthoppers and their close relatives , the cicadas [12 , 24] . Both make use of paired elastic recoil mechanisms and low-frequency active muscle contractions to enhance the efficiency and efficacy of communication , using exoskeletal integration to transform mechanical impulses into substrate vibration [2 , 12] . Driven by a single muscle , the cicadas’ tymbal organs use buckling instability of multiple stiff ribs to store and release elastic energy , turning slow muscle contraction into fast motion as the ribs buckle [12] . Muscle relaxation and the release of energy stored in resilin pads causes the ribs to restraighten again , leading to a second step involving elastic energy release [15] . In contrast , the snapping organ uses two different energy-storage mechanisms for paired elastic recoil: elastic storage in contracted muscle for loading and elastic storage in the deformed exoskeleton for unloading . Instead of buckling like the ribs of a tymbal , the arms of the Y-lobe in the snapping organ use snapping motions similar to those used in fast raptorial strikes by jaws and claws [8 , 9] . Finally , whereas tymbal vibrations in most cicadas are often associated with resonant chambers that act to transform motion into loud acoustic signals [12] , the snapping organ is specialised for substrate-borne vibration generation , with comparable muscle contraction rates that act to transfer mechanical energy into vibrations of the substrate [12] . In conclusion , the unique biomechanics of the snapping organ demonstrate the general importance of elastic recoil mechanisms in the fast motions of small arthropods , extending our knowledge of such mechanisms beyond the simpler one-off ballistic motions that characterise jumping , predatory strikes , and feeding . Elastic recoil is a very general mechanism allowing small animals to overcome the limitations of their size and enabling robust vibrational communication .
Individuals of A . bilobum , the model planthopper species used in this study , were collected in large numbers ( n = 250 ) in late April 2017 as fourth/fifth-instar larvae or adults from Lycabettus Hill , Athens , Greece , and were imported to Oxford , UK under DEFRA Plant Health Licence no . 52972/198417/6 . Larvae were reared into adulthood in mesh cages ( 47 . 5 cm × 47 . 5 cm × 47 . 5 cm ) kept at 22–29°C , 50% humidity , with a 16:8 photoperiod ( light/dark ) . In addition , the morphology of specimens from more than 130 taxa were examined , covering the entire phylogenetic spectrum of Fulgoromorpha . S1 Table details the techniques used to examine the morphology of the snapping organ for each species , along with its preservation method . Planthoppers belonging to 12 families ( including three specimens of A . bilobum: adult male , female , and larva ) were used for synchrotron radiation microcomputed tomography ( SR-μCT ) at the TOMCAT beamline , Swiss Light Source ( SLS ) , Paul Scherrer Institute , Switzerland ( S1 Table ) . All specimens were scanned at a beam energy of 15 . 99 keV with a final pixel size of 1 . 625 μm , allowing visualisation of even the smallest muscles and nerves of the snapping organ ( Figs 1B and 2B and S2 and S3B–S3D ) , which were otherwise not detected by other techniques . Three-dimensional reconstruction was carried out using Amira 6 . 1 software ( Mercury Systems , Andover , MA , USA ) . All shown tomographic data ( reconstructed TIFFs ) for the two imaged species ( A . bilobum and Stenocranus minutus ) are freely available at CXIDB ( http://cxidb . org/id-93 . html ) [35] . Colouration and labelling of figures were performed in Adobe Illustrator CS6 . In order to reveal the primary DVMs operating the snapping organ in A . bilobum , the ventral junction between the Y-lobe and tg2 were excised from an ethanol-preserved ( 70% ) male ( Fig 2C ) . The dissected sample was placed between two cover slips in 70% ethanol and was imaged with a laser confocal scanning microscope ( Olympus FV1000; Olympus , Tokyo , Japan ) at a laser wavelength of 488 nm . The morphologies of specimens belonging to all 21 planthopper families were examined under light microscopy . Images of the snapping organ of four species of planthoppers shown in Fig 3 were taken using a Leica M165c microscope equipped with a Leica DFC490 camera ( Leica , Wetzlar , Germany ) . The final , stacked images were combined using Helicon Focus ( Helicon Soft , Kharkiv , Ukraine ) . Image brightness adjustment was performed in Adobe Photoshop , and drawings were generated in Adobe Illustrator CS6 . To record vibrational signals , planthoppers were placed on a dried grass ( Schedonorus giganteus ) stem ( 17 cm in height ) . The base of the stem was inserted inside an empty c . 1-cm–diameter tube and was held in place by aluminium foil . Vibrational signals were recorded by a laser Doppler vibrometer ( Polytec PDV-100; Polytec , Waldbronn , Germany ) , focussed at different positions approximately orthogonal to the stem and bug . A sampling frequency of 9 . 6 kHz was used for recordings at a gain of 100 mm/s/V . Recording started immediately once the planthoppers were placed on the stem . Each recording lasted 6 minutes and was repeated until the animal either ended its vibrational call or after four recordings if no songs were present . A total of 61 recordings were made , 31 on single planthoppers , 26 on male–female groups , and four on male–male groups , using a total of 19 individuals ( 12 males , 7 females ) . Recordings from two individuals are included in S1 Data , in which the laser was focussed on the plant stem ( individual 1 ) , bug prothorax ( individual 1 ) , bug genitalia ( individual 2 ) , or bug midabdomen ( individual 1 ) . All vibrometry recordings were similar in the type and pattern of motion observed , so the data presented in S1 Data and S5 Fig are assumed to be representative . Attenuation of motion during loading and unloading from the midabdomen to the plant stem and prothorax to the plant stem was calculated in decibels ( S1 Data ) . Vibrometry figures were drawn using Raven Lite 2 . 0 ( Cornell Lab of Ornithology , Ithaca , NY , USA ) and OriginPro 8 . To stimulate vibration generation , we used playback tracks of recorded songs . The stem was vibrated 7 . 3 cm from the base by a pin glued to a small piezo disc ( RS Components , Corby , UK ) , which was glued on an inverted plastic cup . Playback songs consisted of prerecorded and amplified vibrational signals of both sexes . All males responded to the playback by emitting a series of pulses for several minutes . The motion of the snapping organ in A . bilobum was captured with a high-speed camera ( Grasshopper3 2 . 3 MP Colour USB3 Vision , Sony Pregius IMX174; Point Grey , Richmond , BC , Canada ) mounted on a Leica S8 AP0 stereomicroscope , recording at a rate of 100 frames s−1 . Videos were recorded directly to a computer using Spinnaker SDK-1 . 3 . 0 . 21 software ( Point Grey ) . A total of three males were video recorded , and a movie and still frames from one male are given in Fig 4 and S1 Movie . The males were filmed over multiple cycles , frames were classified into the different stages of the mechanism , and the clearest frames were chosen from these classified groups within Fig 4 . Pixel coordinates of three points on the bug prothorax were quantified for each frame used in Fig 4 to check for alignment of the bug within the video frame over time . Standard deviation over the five frames for each of the three points was within the order of 0 . 01 pixels , suggesting the bug has limited movement within the video frame over successive cycles ( also supported by S1 Movie ) . Prior to recording , it was necessary to expose the snapping organ by removing the fore and hind wings with a scalpel . The males were then left on their host plant for one hour to recover after wing removal before playback recordings were started to stimulate vibration generation . Based on our observations , the motion captured in S1 Movie is representative of the vibration-generation mechanism across different individuals . The vibrometry recordings were analysed to calculate the peak energy and power of the loading and unloading motions ( S1 Data ) . Maximum and minimum peak velocities and the timings of the peaks were extracted from the vibrometry data . The peak kinetic energy of the motion was calculated from the speed of the measured dorsoventral translation of the abdominal mass , and the corresponding mechanical power was determined by dividing this peak kinetic energy by the time taken to reach it from rest . The muscle power density that would be required to generate this motion through direct muscle contraction was calculated by dividing these values by the relevant muscle mass , as measured from SR-μCT measurements of A . bilobum , modelling muscles as cylinders with a density of 1 , 060 kg m−3 [36] . A mathematical model was developed to support the interpretation that unloading was due to elastic recoil of the system ( Figs 5 and S4 ) . The model included the abdomen as a mass attached to two rigid bars in series ( anterior Y-lobe arm and ridge , respectively ) , each with a stiff rotational spring at their junctions . The anterior bar was fixed to a surface , representing the thorax . Springs and dampers acting on the mass of the abdomen modelled the combined action of the muscles , resilin , other exoskeletal components , and interior morphology on the motion of the mass in the dorso–ventral and anterior–posterior planes . Full details of the model are given in S1 Methods . | Animals use substrate-borne vibrations for eavesdropping and communication over an immense range of body size—from elephants to nematodes . Vibrational communication is especially challenging for small animals because of the high mechanical power that is needed to transmit information effectively over extended distances through a substrate . Here , we show that planthoppers , a commercially important group of insects , produce vibrations for communication using a reciprocal elastic recoil mechanism that proves remarkably effective at small body size . By combining morphological and biomechanical analyses of a previously overlooked vibratory organ on the abdomen , we show that planthoppers use fast , cyclical abdominal motions to generate substrate-borne vibrations . This novel , to our knowledge , mechanism , which we term the snapping organ , makes use of slow energy storage and fast elastic recoil twice during each cycle of motion , involving two distinct elastic elements . This cyclical mechanism allows planthoppers to transmit signal pulses containing a broad range of frequencies to the substrate . The mechanism is efficient , achieving fast cyclical motion without relying on high muscle power and mass , both of which are limited for animals of small size . The snapping organ is ubiquitous across planthoppers and presents an interesting example of how elastic mechanisms can be used to enable nonacoustic vibrational communication between animals . | [
"Abstract",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"abdomen",
"diagnostic",
"radiology",
"vibration",
"social",
"sciences",
"electromagnetic",
"radiation",
"muscle",
"contraction",
"mechanical",
"energy",
"animal",
"behavior",
"synchrotron",
"radiation",
"zoology",
"research",
"and",
"analysis",
"methods",
"muscle",
"physiology",
"musculoskeletal",
"system",
"animal",
"communication",
"imaging",
"techniques",
"behavior",
"abdominal",
"muscles",
"muscles",
"tomography",
"physics",
"psychology",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
] | 2019 | Planthopper bugs use a fast, cyclic elastic recoil mechanism for effective vibrational communication at small body size |
Kernel row number ( KRN ) is an important component of yield during the domestication and improvement of maize and controlled by quantitative trait loci ( QTL ) . Here , we fine-mapped a major KRN QTL , KRN4 , which can enhance grain productivity by increasing KRN per ear . We found that a ~3-Kb intergenic region about 60 Kb downstream from the SBP-box gene Unbranched3 ( UB3 ) was responsible for quantitative variation in KRN by regulating the level of UB3 expression . Within the 3-Kb region , the 1 . 2-Kb Presence-Absence variant was found to be strongly associated with quantitative variation in KRN in diverse maize inbred lines , and our results suggest that this 1 . 2-Kb transposon-containing insertion is likely responsible for increased KRN . A previously identified A/G SNP ( S35 , also known as Ser220Asn ) in UB3 was also found to be significantly associated with KRN in our association-mapping panel . Although no visible genetic effect of S35 alone could be detected in our linkage mapping population , it was found to genetically interact with the 1 . 2-Kb PAV to modulate KRN . The KRN4 was under strong selection during maize domestication and the favorable allele for the 1 . 2-Kb PAV and S35 has been significantly enriched in modern maize improvement process . The favorable haplotype ( Hap1 ) of 1 . 2-Kb-PAV-S35 was selected during temperate maize improvement , but is still rare in tropical and subtropical maize germplasm . The dissection of the KRN4 locus improves our understanding of the genetic basis of quantitative variation in complex traits in maize .
Understanding the genetic and molecular basis of grain yield is necessary to guide breeding efforts towards the development of high-yielding maize hybrids . Kernel row number ( KRN ) in maize is one of the most important yield components and a significant breeding target . During the domestication of maize , KRN underwent a dramatic change from two rows in teosinte to more than eight rows in modern maize [1] . A number of quantitative trait loci ( QTL ) have been reported [2–3] to control quantitative variation in KRN . However , the genetic and molecular mechanisms of these KRN QTL are unknown . Switching from vegetative to reproductive development turns axillary meristems ( AMs ) into ear inflorescence meristems ( IMs ) [4] . The IMs then elongate and produce spikelet-pair meristems ( SPMs ) . Each SPM makes two spikelet meristems ( SMs ) , which then give rise to floral meristems ( FMs ) that form kernels after fertilization [4] . The initial number of SPMs on the female inflorescence meristem determines the number of kernel rows on the maize ear , while the meristematic activity of IMs determines the potential number of kernels in each kernel row . The initial number of SPMs is correlated with the size of the inflorescence meristem , which provides space for the development of SPMs . The CLAVATA-WUSCHEL ( CLV-WUS ) feedback-signalling loop regulates IM size by restricting stem cell proliferation and maintaining meristem activity . Recently , several genes in the CLV-WUS feedback loop , including thick tassel dwarf1 ( td1 ) [5] , fasciated ear2 ( fea2 ) [6–7] , and COMPACT PLANT2 ( CT2 ) [8] , were isolated in maize . Additionally , the RAMOSA genes [9] , Corngrass1 ( Cg1 ) [10] , tasselsheath4 ( tsh4 ) [11] , FLORICAULA/LEAFY ( ZFL1 and ZFL2 ) [12] , unbranched2 ( ub2 ) and ub3 [13] and others , all affect ear morphology by regulating the development of SPMs and SMs . However , these genes were originally isolated through genetic assays of inflorescence mutants , the mechanisms of them to affect quantitative variation of ear-related traits remain unknown , except for fea2 and ub3 [7 , 13] . Thus , the genetic basis and molecular regulation of quantitative variation in KRN deserves further study . Previously , a major KRN QTL , KRN4 , with a large additive effect was identified by combining linkage and association mapping [2–3] . We found that the associated SNPs within KRN4 constitute a linkage disequilibrium block ( Chr4:198 . 9Mb–199 . 9Mb ) in our association mapping panel ( S1 Fig ) . In the present study , we isolated KRN4 by positional cloning and analysed the putative causal variant using maize mutants , gene expression , and association mapping . We then examined changes in the allelic composition of populations for the causal variant during the domestication and improvement of maize . Finally , we assessed the utility of KRN4 for maize breeding by allele substitution using marker-assisted selection .
To fine-map KRN4 , a near isogenic line ( H21NX531 ) containing the QTL was developed . In comparison with H21 , H21NX531 exhibited similar plant appearance ( Fig 1A ) . The KRN ( P-value = 5 . 87 E-07 ) , ear diameter ( P-value = 0 . 0017 ) , cob diameter ( P-value = 0 . 0075 ) , kernel number ( P-value = 8 . 70 E-05 ) , and grain yield ( P-value = 7 . 47 E-05 ) were significantly increased in H21NX531 ( Table 1 and Fig 1B ) . However , 100-kernel weight of H21NX531 did not differ from that of H21 ( Table 1 ) . To understand the developmental basis of the increase in KRN , we measured the inflorescence meristem size of the 2-mm immature ear . The diameter of ear IM in H21NX531 is significantly larger ( P-value = 5 . 2 E-04 ) than that of H21 in the developing female inflorescence ( Fig 1C and 1D ) . Next , to fine map KRN4 , a total of 31 recombinants representing 13 distinct crossover events were found in over 10 , 000 F2 individuals derived from the cross H21×H21NX531 . We compared the KRN of H21 with homozygous recombinant lines derived from the 13 representative recombinants , and found that the homozygous recombinant lines ( RL2 , RL4 , RL5 , RL6 , RL7 , and RL11 ) carrying the H21NX531 genomic segment between marker M6 and M8 displayed higher KRN ( more than 13 rows , P-value < 1 . 0 E-05 , Student’s t-test ) than H21 ( 11 . 8 ± 1 . 3 ) , while the other homozygous recombinant lines carrying the H21 genomic segment exhibited almost the same KRN as H21 ( Fig 2A ) . To exclude the effect of residual genetic background , we also compared the KRN of offspring individuals derived from each of the 13 heterozygous recombinants in four environments . We found that only when the offspring populations were segregated with KRN4H21 and KRN4NX531 in M6-M8 marker interval ( RL6-RL10 , S1 Dataset ) , the KRN of those individuals with the homozygous H21NX531 genotype in the M6-M8 marker interval were significantly higher than that of individuals with the homozygous H21 genotype ( P-value < 0 . 01 , S1 Dataset , Student’s t-test ) . Therefore , we could narrow the genomic location of KRN4 down to a 3-Kb intergenic region flanked by M6 and M8 markers ( Fig 2A and S1 Dataset ) , which is located ~60 Kb downstream from an SBP-box gene UB3 [13] and ~300 bp upstream of a gene of unknown function , GRMZM2G001541 ( Fig 2A ) . The genomic region between marker M6 and M8 was defined as KRN4 . In comparison with H21 , two regions totaling 1 . 2 Kb in length ( the 1 . 2-Kb PAV ) each containing a fragment of the harbinger transposable element are present in H21NX531 ( Fig 2B and S2 Dataset ) . Several SNPs and small indels are also present in this region ( Fig 2B and S2 Dataset ) . Therefore , sequence differences within the 3-Kb genomic region between H21 and H21NX531 could be the potential causative sites for KRN4 to control KRN variation . We first examined the expression atlas for UB3 and GRMZM2G001541 . The expression data were obtained from qteller ( http://www . qteller . com/ ) and MaizeGDB ( http://www . maizegdb . org/ ) . We found both UB3 and GRMZM2G001541 exhibited similar mRNA expression patterns and accumulated in developing ears and tassels ( S2 Fig ) . They also express in the non-reproductive tissues such as leaf , internode etc . ( S2 Fig ) . However , in the immature ear at spikelet-pair meristems ( 2-mm ear ) and spikelet meristems ( 5-mm ear ) differentiation stages , only UB3 exhibited differential expression between H21 and H21NX531 , with an expression level almost threefold higher in H21 than in H21NX531 ( Fig 3A ) . Differential expression of UB3 was also observed in stems , roots , and leaves ( S3A Fig ) . However , in 5-mm tassel and 10-mm tassel , expression of UB3 did not show an obvious decrease in H21NX531 relative to H21 ( S3A Fig ) , which might explain why tassel branch number did not differ between H21 and H21NX531 ( Table 1 ) . To explore the relationship between expression of UB3 and KRN4 , we analysed the expression of UB3 and GRMZM2G001541 in immature ears of six homozygous recombinant lines ( RL4 , RL5 , RL6 , RL7 , RL8 , and RL12 ) and two parental lines ( H21 and H21NX531 ) , and found that RL4 , RL5 , RL6 , and RL7 , which carry the KRN4NX531 allele , showed lower expression of UB3 and higher KRN , while the lines RL8 and RL12 , which carry the KRN4H21 allele , showed higher expression of UB3 and correspondingly lower KRN ( Fig 3B ) . In contrast , the expression of GRMZM2G001541 in the lines with the KRN4NX531 allele was similar to that in lines with the KRN4H21 allele ( P-value = 0 . 42 ) ( Fig 3B ) . Therefore , the expression of UB3 is regulated by KRN4 , shows a strong negative correlation with KRN ( Fig 3B ) . We further divided these 38 diverse maize inbred lines into two groups: Group L carrying the KRN4H21 allele ( N = 26 ) and Group H carrying the KRN4NX531 allele ( N = 12 ) , according to their genotypes for the 1 . 2-Kb PAV of KRN4 ( S1 Table ) . By examining UB3 expression at the 2-mm ear stage , we found that the expression of UB3 in Group L lines was significantly higher than that in Group H lines ( P-value = 0 . 038 , Student’s t-test , Fig 3C ) , and KRN in these 38 inbred lines was again negatively correlated with the expression level of UB3 ( r = -0 . 35 , P-value = 0 . 037 , Pearson’s correlation coefficient , S3B Fig ) . We sequenced KRN4 ( ~3 Kb , between marker M6 and M8 ) and UB3 genic region ( ~4 Kb , including promoter to 3′-UTR but not first intron ) in our association mapping panel ( S3 Dataset ) [3 , 14] , and identified 69 and 46 polymorphic sites , respectively , with Minor Allele Frequency ( MAF ) ≥ 0 . 05 ( S4 Fig ) . Association analysis using the MLM K + Q model [15–16] revealed that four sites were associated with KRN at P-value <1 . 0 E-04 ( Table 2 ) , including one A/G SNP in the third exon of UB3 ( S35 , P = 3 . 81E-08 , N = 428 ) , one G/A SNP in the 3'-UTR region of UB3 ( S45 , P-value = 7 . 35 E-05 , N = 384 ) , one ~700 bp insertion/deletion ( S23 , P-value = 6 . 69 E-05 , N = 416 ) in the promoter region of UB3 , and the 1 . 2-Kb PAV in KRN4 ( P-value = 7 . 28 E-06 , N = 428 ) ( Table 2 ) . The four sites could be classified into three LD groups at R2 > 0 . 4: group 1 including S23 , group 2 including S35 and S45 , and group 3 including the 1 . 2-Kb PAV ( S4 Fig ) . Conditional association analysis was then conducted using these four sites as covariates under an MLM K + Q model , to determine whether these sites were independent or not . When S35 was conditioned , neither S45 nor S23 were significantly associated with KRN ( P-value 0 . 49 and 0 . 41 , S2 Table ) , but the 1 . 2-Kb PAV was found to be weakly associated with KRN ( P-value = 0 . 03 , S2 Table ) . The signals for association of S35 and the 1 . 2-Kb PAV with KRN were only slightly decreased when conditioned by any one of S23 and S45 ( S2 Table ) . Finally , when conditioned on the 1 . 2-Kb PAV , the other variants were also still significantly associated with KRN ( S2 Table ) . Hence , the association of the 1 . 2-Kb PAV with KRN might be independent of S23 and S45 but partially related to S35 , and the association of S23 and S45 with KRN might depend on that of S35 . The dependence of S45 on S35 might be due to its high linkage disequilibrium with S35; thus , S35 could actually represent the association of S45 with KRN , while S23 might not , because of the weak linkage disequilibrium between S23 and S35 ( R2 = 0 . 21 ) . To further determine the relationship between the 1 . 2-Kb PAV , S35 , and S23 , the segregating populations derived from selfing the heterozygous recombinants RL6-RL12 were used to evaluate the additive effects of these three tightly linked loci . The 1 . 2-Kb PAV showed a large additive effect ( 0 . 78 ) in RL6 offspring segregating population , while the additive effect of S35 and S23 were zero in RL11-RL12 ( Fig 4 ) . However , combination of the 1 . 2-Kb PAV + S35 ( RL7 ) or the 1 . 2-Kb PAV + S35 + S23 ( RL8-RL10 ) had an additive effect more than 1 . 07 rows , almost 40% higher than that of the 1 . 2-Kb PAV alone in RL6 ( Fig 4 ) . These two kinds of combinations exhibited a similar additive effect , which suggests that the increased additive effect was caused mainly by S35 or polymorphisms tagged by S35 . Therefore , the 1 . 2-Kb PAV or a locus near 1 . 2-Kb PAV that genetically interacts with a locus tagged by S35 , and their interaction , might strongly promote the additive effect on KRN ( Fig 4 ) . We next constructed haplotypes using 1 . 2-Kb PAV and S35 ( 1 . 2-Kb-PAV-S35 ) and found that they showed stronger association with KRN ( P-value = 2 . 41 E-09 , N = 428 , MLM K + Q ) than did each individual locus , when comparing the high-KRN haplotype against the low-KRN haplotype using the MLM K + Q model . In the association mapping panel , a total of four haplotypes ( Hap1-Hap4 ) were observed for the 1 . 2-Kb-PAV-S35 ( Table 3 ) . Lines with Hap1 exhibited higher KRN than lines with the other three haplotypes , and lines containing Hap2 to Hap4 did not significantly differ from each other in KRN ( Table 3 ) . A total of 29 maize wild relatives Z . mays subsp . parviglumis teosinte accessions and 36 diverse maize landraces were employed to estimate the selection pressure during maize domestication ( S4 Dataset ) . The genomic sequence of KRN4 was sequenced in them . Then three expectations of past selection were assessed . First , we compared the nucleotide diversity ( π ) of KRN4 between teosintes and maize landraces . We found KRN4 had undergone strong reduction in nucleotide diversity from teosintes to maize landraces with πmaize/πteosinte = 0 . 10 , indicating that only 10% nucleotide diversity in teosintes was retained in maize landraces ( Fig 5A ) . Second , a significantly negative Tajima’s D-statistic ( -2 . 18 , P-value < 0 . 01 , length of tested region = 3 , 144 bp , number of sites = 1 , 722 , Fig 5A ) of KRN4 was acquired in maize landraces which suggested a recent selection in the KRN4 region . Furthermore , the Hudson–Kreitman–Aguade ( HKA ) test was applied to assesses the ratio of diversity in maize landrace to divergence from an outgroup ( Z . diploperennis ) for KRN4 relative to four neutral genes . KRN4 in landrace showed significant selection based on HKA test result ( P-value = 3 . 32E-04 , length of tested region = 3 , 144 bp , number of sites = 1 , 722 , Fig 5A and S3 Table ) , but KRN4 in teosinte doesn’t ( P-value = 0 . 46 , length of tested region = 3 , 300 bp , number of sites = 1 , 642 , Fig 5A and S3 Table ) . These results revealed that KRN4 was under strong selection during domestication from teosinte to maize , similar to tga1 promoter and tb1upstream region [17–18] . However , different from tga1 and tb1 loci [17–18] , no fixed difference between teosintes and maize landraces could be observed in KRN4 . To explore the evolution of 1 . 2-Kb-PAV and S35 loci , we genotyped them in 120 teosinte accessions , 280 maize landraces ( S5 Dataset ) and 428 maize inbred lines , respectively . In teosinte , the frequencies of favorable alleles for 1 . 2-Kb PAV ( 1 . 2-Kb Presence allele ) and S35 ( A allele ) were 5 . 8% and 0% ( Fig 5B and 5C ) . The 1 . 2-Kb Presence allele had a higher frequency ( 9 . 6% ) in Z . parviglumis but rare in Z . mexicana ( 2 . 4% ) , implying that the favorable allele of 1 . 2-Kb PAV in modern maize was probably selected from Z . parviglumis ( S5 Dataset ) . In maize landrace , the frequencies of favorable alleles for 1 . 2-Kb PAV and S35 were increased to 18 . 6% and 1 . 25% , respectively ( Fig 5B and 5C ) . During modern maize improvement , they were enriched to 36 . 1% and 13 . 2% ( Fig 5B and 5C ) , and the R2 of them in the association mapping panel were 5 . 0% and 12 . 2% . The favorable haplotype of 1 . 2-Kb-PAV-S35 , Hap1 was not detected in teosinte accessions ( Fig 5D ) , and the frequency of Hap1 in maize inbred lines increased to 12 . 8% ( N = 428 , Fig 5D ) , but differed dramatically between temperate ( 21 . 1% , N = 234 , Fig 5D ) and TST ( tropical and subtropical , 2 . 5% , N = 194 , Fig 5D ) maize inbred lines . The unequal distribution of Hap1 in different subpopulations suggests that favorable Hap1 has been selected to increase grain yields by increasing the number of kernel rows in temperate germplasm . Based on these results , we proposed an evolutionary pattern of 1 . 2-Kb PAV and S35 during maize domestication and improvement ( Fig 5D ) . Hap2 of 1 . 2-Kb-PAV-S35 , which harbors the 1 . 2-Kb Presence allele , was selected and enriched from teosinte to landrace and then to tropical and subtropical maize inbred lines ( Fig 5D ) . The favorable Hap1 allele might have been selected from teosinte or could have arisen by mutation at S35 after domestication ( Fig 5D ) . However , the intensive selection on Hap1 only occurred during temperate maize inbred lines improvement ( Fig 5D ) . UB3 is an ortholog of OsSPL14 , which is responsible for IPA1 ( ideal plant architecture 1 ) and WFP ( WEALTHY FARMER’S PANICLE ) in rice ( S5 Fig ) [19–20] , and is also homologous with UB2 . Recent study has revealed that ub2 and ub3 knock-out mutants exhibit increase in maize KRN [13] . Two novel Mutator-mediated mutants , UB3-mum4 , with a Mu7 insertion in the promoter region of UB3 , and UB2-mum3 , with a Mu7 insertion in the first intron of UB2 ( Fig 6A ) , were obtained from Maize Stock Center . UB3 expression in 2-mm immature ears and 5-mm tassels of the UB3-mum4 line was significantly higher than that in the wild type ( WT ) ( Fig 6B ) . Similarly , a previous study has identified that a Mu transposon insertion in 5’UTR of P1 gene increases P1 expression in maize [21] . UB2 expression in 2-mm immature ears of the UB2-mum3 line did not differ significantly from WT ( Fig 6C ) , but ~14% of UB2-mum3 transcripts contained an extra 295-bp fragment composed of a 145-bp intron sequence flanking Mu7 insertion sites and a 150-bp terminal inverted repeat of Mu7 ( S6 Fig ) . The 295-bp fragment was inserted into the SBP-box domain-encoding sequence and might result in loss of function of the alternatively spliced transcript . We developed segregating populations to evaluate the influence in KRN by the Mu7 insertion in UB3-mum4 and UB2-mum3 . Each single mutant did not show an obvious change in KRN or ear diameter ( Fig 6D and S4 and S5 Tables ) , only UB3-mum4 showed a slight but significant decrease in KRN in 2013 Wuhan environment ( P-value = 0 . 01 , Fig 6D and S4 Table ) . Interestingly , double mutants of UB3-mum4 and UB2-mum3 showed a significant decrease in KRN ( P-value = 2 . 21 E-04 ) and ear diameter ( P-value = 2 . 90 E-05 ) relative to WT ( Fig 5D and 5E and S6 Table ) . In addition , UB3-mum4 and double mutant also showed a slight but significant reduction in tassel branch number relative to wild types ( Fig 6E and S6 Table ) . The introgression of the 1 . 2-Kb PAV from NX531 into H21 results in significant enlargement of the inflorescence meristem in the immature ear of H21NX531 ( Table 1 and Fig 1C and 1D ) . The enlarged diameter of the inflorescence meristem provides a larger space to support the larger number of spikelet-pair meristems generated . Accompanying the increase in KRN in H21NX531 , kernel number per ear also significantly increased , but 100-kernel weight was not affected , and so the grain yield of H21NX531 was markedly enhanced ( Table 1 ) . The enhanced yield resulting from the increased KRN with unaltered kernel weight may only apply to the specific genetic backgrounds or growth conditions . Then , we anticipate that selection for the favorable allele at KRN4 will contribute positively to maize productivity . To test this hypothesis , we used marker-assisted selection to introgress the 1 . 2-Kb Presence alleles from two inbred lines carrying the 1 . 2-Kb Presence alleles , TY6 and Qi205 , into W138 and Mo17 carrying the 1 . 2-Kb Absence alleles . To minimize the influence of genetic background , heterozygotes at the 1 . 2-Kb PAV in BC3F1 were selfed to develop a segregating population , and then two homozygous genotype subgroups ( 1 . 2-Kb Presence subgroup , 1 . 2-Kb Absence subgroup ) were identified in each segregating population for KRN evaluation to maximum randomize genetic background . We found that mean of KRN of the 1 . 2-Kb Presence subgroup was almost 2 rows higher than that of the 1 . 2-Kb Absence subgroup , indicating that the introgression of the superior alleles could increase KRN of recurrent parents ( S7 Table ) .
In this study , we fined mapping a major KRN QTL , KRN4 , and suggested the 3-Kb intergenic region that includes a 1 . 2-Kb PAV ~60 Kb downstream of UB3 is the causation underlying the major KRN QTL . Expression analysis in immature ear indicated that the expression difference of UB3 between H21 and H21NX531 , and also among diverse inbred lines , was highly correlated with variation in KRN4 . Further , the weak mutants of UB3-mum4 and UB2-mum3 used in this study demonstrated that elevation of UB3 expression reduces the KRN and ear diameter , which is consistent with previous characterized ub3 and ub2 knock-out mutations which cause KRN increase and ear diameter enlargement [13] . The elevation of UB3 expression in UB3-mum4/UB2-mum3 may reduce the inflorescence meristem size of the developing ear , resulting in formation of less spikelet-paired meristems ( SPMs ) , and then decreased number of kernel rows and ear diameter . This hypothesis can be supported in H21 and H21NX531 , where the higher UB3 expression in H21 is correlated with smaller inflorescence meristem size and less SPMs formation than H21NX531 , and also is consistent with ub3 knock-out mutants with enlargement in inflorescence meristem size [13] . However , we observed that an increase of UB3 expression in UB3-mum4 slightly reduces the tassel branch number , which is inconsistent with the results of ub3 knock-out mutants , which show highly suppressed tassel branch [13] . These observations imply that the allele effect on tassel branch number of UB3-mum4 used in this study is different from previous identified ub3 knock-out mutants . The ortholog of UB3 and UB2 in rice , OsSPL14 , negatively regulates axillary bud outgrowth to repress shoot tillering , but positively regulates the number of panicle branches by enhancing meristematic activity and cell proliferation [19–20 , 22–24] . Unlike OsSPL14 , UB3 and UB2 exhibit redundant biological functions on negative regulation of KRN , a kind of short branch in maize ear . It seems like that UB3 and UB2 evolved from a common ancestral gene with OsSPL14 and retained similar biological functions , but may act in opposite ways . Therefore , we suggest that KRN4 controls the natural variation of KRN by acting as a distal regulator of UB3 expression and UB3 negatively regulates KRN in maize . Previous study revealed that ub3 shows more severe phenotype than ub2 [13] . The UB3 locus is also a KRN and tassel branch number QTLs hotspot detected by many studies [2–3] , and UB3 is found to be the causative gene underlying a major KRN QTL , KRN4 , in this study . However , the natural variation in UB2 locus has not been found to be associated with inflorescence traits in maize [2–3] . So , alterations in UB3 by mutations or natural variation are more likely to cause the response on inflorescence traits than UB2 . In addition , the expression differences of UB3 was not in developing tassels , consistent with ear traits being modulated and tassel traits not . Thus , KRN4 may not be responsible for the TBN QTLs at this locus , which is consistent with previous suggestion that KRN and TBN are controlled by different polymorphisms of UB3 [13] . The association analysis of KRN4 revealed that only the 1 . 2-Kb PAV containing TE fragments was significantly associated with KRN in diverse inbred lines . Hence , variation in KRN between H21 and H21NX531 due to UB3 expression is possibly caused by the 1 . 2-Kb PAV . This kind of distal regulation of gene expression being responsible for variation in important traits has been previously described in maize , and two different mechanisms may account for it . First , like tb1 , Vgt1 , ZmCCT , and prol1 . 1 , the causal sequences ( commonly transposon derived sequences ) act as enhancers to regulate gene expression level or pattern in cis [18 , 25–28] . In a second mechanism , non-coding tandem repeat sequences located ∼100 kb upstream of b1 express dsRNA , which mediates trans-communication between alleles to establish paramutation [29] . KRN4 may interact with the UB3 regulatory region in cis to promote expression of UB3 , or the transposon fragments in KRN4 may express small RNAs and affect UB3 expression by an epigenetic regulation mediated by small RNAs . These assumptions are yet to be investigated . In addition to 1 . 2-Kb PAV , an A/G SNP designated as S35 that is significantly associated with KRN was also detected in our association mapping panel . Located in an exon of UB3 , this is the same as the Ser220Asn polymorphism mentioned by Chuck et al . [13] . S35 showed stronger association with KRN and had better support in conditional analysis than did the 1 . 2-Kb PAV . However , unlike the 1 . 2-Kb PAV , in the recombinant lines of the fine mapping population , the introgression of A ( or Asn220 ) from H21NX531 to replace the G ( or Ser220 ) in H21 did not result in increased KRN in RL12 . Further , when S35 was segregating in RL11-RL12 , no significant additive effect was observed . But the additive effects of 1 . 2-Kb PAV could be promoted 40% by S35 in the background of 1 . 2-Kb PAV , implying a positive genetic interaction between them and a larger genetic effect due to their combination . This hypothesis is supported by the stronger association of KRN with the creating haplotype 1 . 2-Kb-PAV-S35 than with either of the individual loci . We propose that a change in UB3 protein function due to S35 made UB3 more efficient in modulating inflorescence development . Although S35 alone or other polymorphisms in linkage disequilibrium with KRN4 did not display apparent genetic effects in H21 , S35 might still affect the biological function of UB3 in KRN formation in another genetic background . Therefore , the 1 . 2-Kb-PAV-S35 combination could represent the high- and low-KRN haplotypes for KRN4 among these diverse inbred lines , and Hap1 was the most favorable haplotype for KRN . Domestication leads to the loss of genetic diversity throughout the genome , or in specific regions , and desirable alleles for important traits have been selected and enriched [17–18 , 30] . For KRN4 , the nucleotide diversity in maize landrace is markedly reduced relative to that in teosinte . The strong selection signal was also observed by Tajima’s D test and HKA test . Accompanying the selection on KRN4 , the 1 . 2-Kb Presence allele was continuously enriched during maize domestication and improvement for desirable alleles of KRN4 . Its frequency was increased more than twofold from teosinte to maize landrace , and was further doubled from landrace to modern inbred line . In the corresponding processes , the mean values of allele frequency at four neutral genes ( adh1 , adh2 , fus6 and te1 ) were small changed , just 0 . 37 fold change from teosinte to landrace , and 0 . 15 fold change from landrace to modern inbred line for low frequent allele , respectively . Additionally , the favourable allele of KRN4 was enriched rather than was fixed in modern maize lines , which is different from the case of tga1 ant tb1 , indicating that KRN4 may be not the critical locus that determines the transition from 2 rows in teosinte to more than 4 rows in modern maize . This was further supported by the fact that neither KRN4 nor UB3 is located within domestication-associated QTL [30] . However , the favourable A allele of S35 in UB3 is not detected in teosinte and has low frequency in maize landraces , indicating that it might have emerged during the post-domestication improvement of modern maize . Because of the larger genetic effect exhibited by the interaction between 1 . 2-Kb Presence allele of KRN4 and A allele of S35 , Hap1 was likely the selection target in modern temperate maize improvement , and the frequency of Hap1 increased more than 7 folds from tropical to temperate maize . Meanwhile , the frequency of A allele of S35 is enriched in temperate maize , but the 1 . 2-Kb Presence allele shows similar frequency between tropical and temperate maize . The decrease of selection pressure on KRN4 during temperate maize breeding might be caused by the selection on the other KRN loci or the diverse breeding objectives . Despite the continued improvement during breeding program , the favourable Hap1 is still absent in most modern maize inbred lines that are included in our association mapping panel . For the TST lines in our association mapping panel , Hap1 was still a rare haplotype . Thus KRN4 and UB3 could be subjected to more intense selection by molecular breeding to improve yield by increasing number of kernel rows in maize ear . In conclusion , the dissection of KRN4 in our study not only extends our knowledge about the genetic and molecular mechanisms of important traits in maize , but also provides diagnostic and germplasm tools for improving maize KRN and grain yields .
A subset of an association mapping panel with 368 diverse inbred lines was genotyped with 500K SNP markers [31] . KRN of these 368 lines was evaluated in five environments and reported in previous study , including Ya'an ( 30°N , 103°E ) , Sanya ( 18°N , 109°E ) , and Kunming ( 25°N , 102°E ) in 2009 , and Wuhan ( 30°N , 114°E ) and Kunming ( 25°N , 102°E ) in 2010 [3] . The best linear unbiased prediction ( BLUP ) of KRN was estimated using a linear mixed model in SAS software ( SAS Institute Inc . , 2001 ) by previous study [3 , 32] . The association of KRN4 with KRN ( BLUP data ) [3] was established using Tassel v3 . 0 with a mixed linear model ( MLM ) approach considering varietal relatedness ( K ) and population structure ( Q ) ( MLM K + Q ) [3 , 15–16] . The linkage disequilibrium among associated SNPs was estimated using Haploview v4 . 1 [33] . A near-isogenic line , H21NX531 , that incorporates the KRN4 QTL for kernel row number ( Chr4:198 . 9Mb-199 . 9Mb , B73 RefGen V2 , S1 Fig ) , was developed by four cycles of backcrossing ( BC ) followed by two cycles of selfing , using H21 as the recurrent parent and NX531 as the donor of the favorable allele . Over 10 , 000 F2 individuals derived from the H21×H21NX531 cross were genotyped with markers flanking KRN4 and 14 newly developed markers ( Primers were listed in S6 Dataset ) within the QTL interval to identify the recombinants . The heterozygous recombinants were self-crossed to segregate the homozygous recombinant ( HR ) and non-recombinant ( HNR ) progeny pairs from each recombinant derived family . The HR and NHR progeny pairs were phenotyped at Wuhan ( 30°N , 114°E ) and Sanya ( 18°N , 109°E ) in 2013 ( S1 Dataset ) , with two replications under a randomized block design for each . And the HRs and HNRs were self-crossed to generate homozygous progeny lines for replicated testing at Wuhan and Baoding ( 38°N , 115°E ) in 2014 ( S1 Dataset ) with two replications under a randomized block design for each . The substitution mapping procedure widely used in fine mapping [34] was employed by examining the KRN differences between HRs and H21 , also between HRs and HNRs progeny pairs from each recombinant derived family , using Student’s t-test with significant threshold P-value < 0 . 01 . To identify candidate genes for the KRN4 QTL , analysis of the expression of genes in the relevant interval was performed on developing ears and tassels from H21 , H21NX531 , recombinant lines , and 38 diverse inbred maize lines ( S1 Table ) using Quantitative PCR ( qPCR ) . Total RNA was extracted using TRIzol Reagent ( Life Technologies , Invitrogen , Carlsbad , CA , USA ) . Total RNAs of H21 and H21NX531 lines were extracted from roots , leaves , stems , immature 5-mm tassel ( 5-mm tassel , 6-leaf stage with branch meristem initiation ) , immature 10-mm tassel ( 10-mm tassel , 10-leaf stage with branches ) , immature ear stage 1 ( 2-mm ear , 10-leaf stage with Inflorescence meristems IMs and spikelet-pair meristems SPMs ) , and immature ear stage 2 ( 5-mm ear , 12-leaf stage with IM , SPM , and spikelet-meristems SM ) . Total RNAs of 38 diverse maize inbred lines were extracted from immature ears at the S1 stage ( S1 Table ) . Total RNAs of UB3-mum4 and UB2-mum3 lines were extracted from immature 5-mm tassel and 2-mm ear , respectively . DNase I ( TaKaRa Biotech , Dalian , China ) was used to remove genomic DNA contamination . An oligo ( dT ) primer and M-MLV reverse transcriptase ( Invitrogen , Carlsbad , CA , USA ) were used to synthesize first-strand cDNAs . A SYBR Green RT-PCR kit ( Bio-Rad , Hercules CA , USA ) was used to perform qPCR with gene-specific primers ( S7 Dataset ) . Expression levels were normalized using beta-actin ( NM_001155179 ) as an endogenous control . The expression data for UB3 and GRMZM2G001541 in B73 were downloaded from the qTeller website ( www . qteller . com ) and MaizeGDB website ( www . maizegdb . org ) . Two Mutator-mediated insertion mutants were obtained from the Maize Genetics Cooperation Stock Center at the University of Illinois , Champaign-Urbana . According to information from the Maize Stock Center [35] , UB3-mum4 ( UFMu-06293 ) has Mutator ( Mu ) inserted upstream of UB3 , and UB2-mum3 ( UFMu-06514 ) has Mu inserted into the first intron of UB2 . The insertion site of Mu was detected by PCR with gene-specific primers and TIR6 primers designed from the TIR sequence of Mu ( S3 Dataset ) . To characterize the phenotypic effects of the mutants and eliminate the influence of the other Mu insertion , UB3-mum4 and UB2-mum3 were backcrossed with its parent W22 , and self-crossed to develop the F2 segregating populations . In each segregating population , wild types ( +/+ ) and homozygous mutants ( -/- ) were identified by genotyping ( Primers used are listed in S3 Dataset ) and were phenotyped . The UB3-mum4 and UB2-mum3 were crossed to develop double mutant , which was also crossed with W22 , and self-crossed to develop F2 segregating populations . In the segregating populations , double mutant and wild type individuals were genotyped and phenotyped . Student's t-test was used to evaluate the phenotypic differences between wild types and mutants . To discover DNA sequence variation and putative causal polymorphisms in UB3 and KRN4 , gene-specific primers ( S3 Dataset ) were designed to amplify UB3 and KRN4 in the association mapping panel [3 , 14] . We genotyped 428 inbred lines using the 1 . 2-Kb PAV in KRN4 as a marker , and sequenced about 4 . 0 Kb of DNA from 5'-upstream of UB3 to its 3'-UTR and ~3 Kb containing KRN4 in 110 or 428 inbred lines of the AM panel , respectively ( the line number of the lines that were sequenced is listed in the S3 Dataset ) . SNPs and indels with MAF > 0 . 05 were used to estimate pairwise LD and to evaluate the association between polymorphic sites and KRN under the MLM K+ Q model [15–16] . Conditional analysis was conducted using the associated sites as covariates under an MLM K + Q model in Tassel v3 . 0 . The MLM K + Q model was also used for haplotype-based association analysis . The selfed progeny of heterozygous recombinants RL6-RL12 which are segregating at 1 . 2-Kb PAV , S35 and S23 were employed to evaluate the genetic effect of 1 . 2-Kb PAV , S35 and S23 ( S1 Dataset ) . The individuals , which harbored homozygous alleles of the three sites , were used to estimate the additive effects of in each segregating population ( S1 Dataset ) . The selection pressure on KRN4 during the domestication and improvement of maize was estimated using 36 randomly selected landraces ( S4 Dataset ) from 280 diverse maize landrace collections ( S5 Dataset ) [36] and 29 Z . mays subsp . parviglumis teosinte ( S4 Dataset ) from 120 teosinte accessions ( S5 Dataset ) . The KRN4 genomic region was amplified and sequenced using primers listed in S3 Dataset . Nucleotide diversity ( π ) and Tajima’s D were estimated using DnaSP ver . 5 . 0 [37] . The 1 . 2-Kb PAV was treated as single PAV when estimating the nucleotide diversity ( π ) . Four neutral loci ( adh1 , adh2 , fus6 and te1 ) [38–41] were used as controls for the HKA test [42] using Zea diploperennis as the outgroup . The overall HKA P-value was obtained by summing the individual χ2 values of the four control genes . Another 88 teosinte accessions ( including 35 Z . mays subsp . parviglumis and 54 Z . mays subsp . mexicana accessions , S5 Dataset ) and 244 maize landraces were genotyped by a PCR marker for the 1 . 2-Kb PAV and a KASP marker ( http://www . kbioscience . co . uk/ ) for S35 , to estimate their frequency in teosinte accessions and maize landraces ( Primers are listed in S3 Dataset ) . All of the sequences have been deposited in NCBI Genebank KT928654—KT931615 . A total of 130 SBP-box genes were predicted in six plant species , including 16 SBP-box genes from Arabidopsis , 18 from Brachypodium , 18 from sorghum , 19 from rice , 20 from foxtail millet , and 29 from maize [43–48] , and used for phylogenetic analysis . | Maize ( Zea mays L . ) is one of the world's most important sources of calories for humans . With an expanding global population , the demands for maize-derived food , feed , and fuel are rapidly increasing . To meet these needs , geneticists and breeders are facing the challenge of enhancing grain yield through genetic improvement of maize germplasm . Understanding the genetic basis of grain yield is necessary to guide breeding efforts towards the development of high-yielding hybrids . Kernel row number ( KRN ) in maize is one of the most important yield components and a significant breeding target . Over the last few decades , many genes that determine inflorescence development and architecture have been identified and characterized . The formation of kernel rows is an integral part of the development of the female inflorescence in maize . Nevertheless , the genetic basis and molecular regulation of quantitative variation in KRN is poorly understood . This study provides experimental evidence for the hypothesis that variation in intergenic regions can regulate quantitative variation of important grain yield-related traits , and also provides tools for improving KRN in maize . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | KRN4 Controls Quantitative Variation in Maize Kernel Row Number |
Among Chagas disease triatomine vectors , the largest genus , Triatoma , includes species of high public health interest . Triatoma dimidiata , the main vector throughout Central America and up to Ecuador , presents extensive phenotypic , genotypic , and behavioral diversity in sylvatic , peridomestic and domestic habitats , and non-domiciliated populations acting as reinfestation sources . DNA sequence analyses , phylogenetic reconstruction methods , and genetic variation approaches are combined to investigate the haplotype profiling , genetic polymorphism , phylogeography , and evolutionary trends of T . dimidiata and its closest relatives within Triatoma . This is the largest interpopulational analysis performed on a triatomine species so far . Triatomines from Mexico , Guatemala , Honduras , Nicaragua , Panama , Cuba , Colombia , Ecuador , and Brazil were used . Triatoma dimidiata populations follow different evolutionary divergences in which geographical isolation appears to have had an important influence . A southern Mexican–northern Guatemalan ancestral form gave rise to two main clades . One clade remained confined to the Yucatan peninsula and northern parts of Chiapas State , Guatemala , and Honduras , with extant descendants deserving specific status . Within the second clade , extant subspecies diversity was shaped by adaptive radiation derived from Guatemalan ancestral populations . Central American populations correspond to subspecies T . d . dimidiata . A southern spread into Panama and Colombia gave the T . d . capitata forms , and a northwestern spread rising from Guatemala into Mexico gave the T . d . maculipennis forms . Triatoma hegneri appears as a subspecific insular form . The comparison with very numerous Triatoma species allows us to reach highly supported conclusions not only about T . dimidiata , but also on different , important Triatoma species groupings and their evolution . The very large intraspecific genetic variability found in T . dimidiata sensu lato has never been detected in a triatomine species before . The distinction between the five different taxa furnishes a new frame for future analyses of the different vector transmission capacities and epidemiological characteristics of Chagas disease . Results indicate that T . dimidiata will offer problems for control , although dwelling insecticide spraying might be successful against introduced populations in Ecuador .
American trypanosomiasis or Chagas disease is widespread in Latin America from Mexico to Chile and southern Argentina . Although present estimates of 10 to 12 million people infected with the haemoflagellate protozoan species Trypanosoma cruzi represent 6–8 million fewer cases than those reported in the 1980s [1] , it remains one of the most serious parasitic diseases of the Americas for its social and economic impact [2] . Although it can also be transmitted by blood transfusion or across the placenta from infected mothers , most human contamination is attributed to insect vectors in poor rural or periurban areas of Central and South America [1] . Chagas disease vectors are haematophagous reduviid ( Hemiptera: Heteroptera ) insects belonging to the subfamily Triatominae . Species of Triatominae are usually grouped into 17 genera forming five tribes , although other arrangements have been proposed . Of these , Alberproseniini , Bolboderini , Cavernicolini and Rhodniini are considered monophyletic , whereas Triatomini is considered polyphyletic [3] . Among the latter , most of the species ( over 70 ) are included in the genus Triatoma , among which two main clades appear in ribosomal DNA ( rDNA ) sequence phylogenies , corresponding to species of North and Central America and species of South America separated prior to the closing of the isthmus of Panama about 3 million years ago [4]–[6] . Moreover , Triatoma species are distributed in three main groupings: the Rubrofasciata group ( mainly North American and Old World species ) , the Phyllosoma group ( mainly Mesoamerican and Caribbean ) , and the Infestans group ( mainly South American ) , each including different complexes and subcomplexes in a classification which is progressively updated according to new genetic and morphometric data [7] . A priori , all of the over 130 species currently recognized within Triatominae seem capable of transmitting T . cruzi . Among the species of greatest epidemiological significance as domestic vectors , three belong to the genus Triatoma: T . infestans and T . brasiliensis from South America , and T . dimidiata , distributed in Meso- and Central America from Mexico down to Colombia , Venezuela , Ecuador and northern Peru [3] . Triatoma dimidiata can be found in sylvatic , peridomestic and domestic habitats . Non-domiciliated populations may act as reinfestation sources and become involved in the transmission of the parasite to humans [8] , [9] . This species includes morphologically variable populations [10] , [11] . A molecular comparison of Triatominae , including many Central American species of the Phyllosoma complex by means of rDNA second internal transcribed spacer ( ITS-2 ) sequences demonstrated an unusual intraspecific sequence variability in a few T . dimidiata populations studied . This study even revealed differences consistent with a specific status for populations from the Yucatan peninsula , Mexico [4]–[6] , thus opening a debate . A large number of recent , multidisciplinary studies using RAPD-PCR , genital structures , morphometrics of head characters , and antennal phenotypes have shown that variation within this species seems much greater than previously considered [8] , [12]–[16] . Morphometric and cuticular hydrocarbon analyses suggest that a sylvatic population from Lanquin , Guatemala , is undergoing a speciation process [13] , [17] . Chromosomal variation and genome size suggest that T . dimidiata may represent a complex of cryptic species ( i . e . morphologically indistinguishable , yet reproductively isolated taxa ) [18] . The aim of the present work is to analyze the intraspecific variability , haplotype profiling , phylogeography and genetic polymorphism of populations of the species T . dimidiata , to get a new framework able to facilitate the future understanding of the diferring peculiarities of this crucial vector species throughout its broad geographical distribution . This may also help in understanding the related differences in characteristics of Chagas disease transmission and epidemiology , as well as in responses to control initiatives in the countries concerned . After a deep analysis , it was considered that the most convenient approach would be obtained by using an appropriate marker able to furnish significant information about evolutionary trends of variation on which to construct the new baseline . This new baseline should be , whenever possible , of sufficient weight as to allow its conclusions to be reflected at systematic-taxonomic level . For this purpose , the rDNA was preferred over mitochondrial DNA ( mtDNA ) because of its mendelian inheritance , evolutionary rates and overall recognized usefulness in systematics in all metazoan organism groups because of including sequences which allow to distinguish between species and between subspecies units . The better fitting of rDNA for molecular systematics has already been emphasized in large reviews on rDNA/mtDNA marker comparisons in insects [19] . Ribosomal DNA includes excellent genetic markers , because ( i ) the rDNA operon is tandemly repeated and present in sufficiently high quantities among the genome of an individual thus facilitating sequencing procedures; ( ii ) the different genes and spacers of the rDNA follow a concerted evolution which , with sufficient time , effectively homologizes the many copies of nuclear rDNA within a genome [20]; this gives rise to a uniformity of their sequences within all individuals of a population and becomes extremely useful from an applied point of view , because it is sufficient to obtain the sequence of only one individual to characterize the local population it belongs to , that is , all other individuals of that population will present the same sequence; ( iii ) the usefulness of rDNA genes and spacers as genetic markers at different evolutionary levels have already been verified on a large number of very different eukaryotic organism groups including insects , and consequently extensive knowledge on the different rDNA fragments is available [21] . rDNA sequence comparisons offer valuable information about the evolutionary events in triatomine lineages and , by deducing the routes of spreading of triatomine populations , they may also shed light on the ability of different species to colonize new areas [5] . Within rDNA , ITS-2 was selected as marker because of its well-known usefulness at species and subspecies levels , including the differentiation of taxa within problematic groups , as is the case of those comprising cryptic or sibling species of other insect groups [22]–[24] . Moreover , the sequences of the ITS-2 have already proved to be a useful tool in the analysis of species , subspecies , hybrids and populations , and for inferring phylogenetic relationships in Triatominae in general [4] , [5] , [6] , [25] , [26] . In order to be able to assess the ITS-2 evolutionary processes followed by T . dimidiata populations , the ITS-2 sequences of many members of the Phyllosoma , Rubrofasciata and Infestans groups were obtained and analyzed . For this purpose , a large number of rDNA ITS-2 sequences of Triatoma species from numerous geographic origins in Mexico , Guatemala , Honduras , Nicaragua , Panama , Cuba , Colombia , Ecuador , and Brazil was studied . Thus , the nucleotide divergence limits between taxa within the lineage of the genus Triatoma could be established . The present study on T . dimidiata is the largest interpopulational analysis performed on a triatomine species so far .
A total of 165 triatomine specimens representing 13 Triatoma species of the Phyllosoma , Rubrofasciata and Infestans groups , among which 137 specimens representing T . dimidiata from 64 different geographic origins , were used for sequencing , genetic variation and phylogenetic analyses ( Table 1; Figure 1 ) . The systematic classification recently proposed for the genus Triatoma [7] is used here throughout . For DNA extraction , one or two legs fixed in ethanol 70% from each specimen were used and processed individually , as previously described [5] , [27] . Total DNA was isolated by standard techniques [28] and stored at −20°C until use . The complete ITS-2 fragment was PCR amplified using 4–6 µl of genomic DNA for each 50 µl reaction . Amplifications were generated in a Peltier thermal cycler ( MJ Research , Watertown , MA , USA ) , by 30 cycles of 30 sec at 94°C , 30 sec at 50°C and 1 min at 72°C , preceded by 30 sec at 94°C and followed by 7 min at 72°C . PCR products were purified with Ultra Clean™ PCR Clean-up DNA Purification System ( MoBio , Solana Beach , CA , USA ) according to the manufacturer's protocol and resuspended in 50 µl of 10 mM TE buffer ( 10 mM Tris-HCl , 1 mM EDTA , pH 7 . 6 ) . Sequencing was performed on both strands by the dideoxy chain-termination method , and with the Taq dye-terminator chemistry kit for ABI 3730 and ABI 3700 capillary system ( Perkin Elmer , Foster City , CA , USA ) , using the same amplification PCR primers [6] . The haplotype ( H ) terminology used in the present paper follows the nomenclature for composite haplotyping ( CH ) recently proposed [25] . Accordingly , ITS-2 haplotypes ( H ) are noted by numbers ( Table 1 ) . Sequences were aligned using CLUSTAL-W version 1 . 83 [29] and MEGA 3 . 1 [30] , and assembly was made with the Staden Package [31] . The alignment was carried out using the Central , Meso and South American Triatoma species studied together with other species and populations whose sequences are available in GenBank: T . phyllosoma ( Accession Number AJ286881 ) , T . pallidipennis ( AJ286882 ) , T . longipennis ( AJ286883 ) , T . picturata ( AJ286884 ) , and T . mazzotti ( AJ286885 ) ( Phyllosoma group , Phyllosoma complex ) ; T . barberi ( AJ293590 ) ( Rubrofasciata group , Protracta complex ) [5] , [6]; T . rubrovaria H1 ( AJ557258 ) [32] , T . infestans CH1A ( AJ576051 ) , and T . sordida ( AJ576063 ) [25] . The ITS-2 sequence of Rhodnius prolixus ( Triatominae: Rhodniini ) ( AJ286882 ) [6] was used as outgroup . The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession numbers for the new ITS-2 rDNA sequences discussed in this paper are: 31 haplotypes of T . dimidiata ( AM286693–AM286723 ) , T . bassolssae AM286724 , T . bolivari ( AM286725 ) , 2 haplotypes of T . hegneri ( AM286726 , AM286727 ) , T . mexicana ( AM286728 ) , 2 haplotypes of T . pallidipennis ( AM286729 , AM286730 ) , T . ryckmani ( AM286731 ) , T . flavida ( AM286732 ) , T . gerstaeckeri ( AM286734 ) , T . rubida ( AM286735 ) , T . nitida ( AM286733 ) , T . maculata ( AJ582027 ) , and T . arthurneivai ( AM286736 ) . Phylogenies were inferred by maximum-likelihood ( ML ) using PAUP*4 . 0b10 [33] and PHYMLv2 . 4 . 4 [34] . Maximum-likelihood parameters and the evolutionary model were determined using the hierarchical Likelihood Ratio Test ( hLRTs ) and the Akaike Information Criterion ( AIC ) [35] , [36] implemented in Modeltest 3 . 7 [37] in conjunction with PAUP*4b10 . To assess the reliability of the nodes in the ML tree , a bootstrap analysis using 1000 pseudo-replicates was made with PHYML . Since haplotype sequences for T . dimidiata individuals ( populations ) are quite similar and potentially subject to homoplasy and recombination , alternative procedures to phylogenetic tree reconstruction revealing their relationships were tested . Therefore , a median-joining network analysis [38] was performed using Network version 4 . 1 . 1 . 2 ( available from Fluxus Technology Ltd . , http://www . fluxus-engineering . com ) with the variable positions in the multiple alignment of the different ITS-2 haplotypes from T . dimidiata populations . Alternative methods of phylogenetic reconstruction allowing an evaluation of the support for each node were also applied . A distance-based phylogeny using the neighbor-joining ( NJ ) algorithm [39] with the ML pairwise distances was obtained . Statistical support for the nodes was evaluated with 1000 bootstrap replicates , with and without removal of gapped positions . Finally , a Bayesian phylogeny reconstruction procedure was applied to obtain posterior probabilities ( BPP ) for the nodes in the ML tree . We used the same evolutionary model as above implemented in MrBayes 3 . 1 [40] with four chains during 1 , 000 , 000 generations and trees were sampled every 100 generations . The last 9 , 000 trees were used to obtain the consensus tree and posterior probabilities . Genetic variation within and among populations of T . dimidiata was evaluated using DnaSP version 4 [41] and Arlequin 2000 [42] . Summary parameters include those based on the frequency of variants ( haplotype number and diversity ) as well as some taking genetic differences among variants into account ( gene diversity , polymorphic sites ) . A hierarchical analysis of molecular variance ( AMOVA ) was performed using Arlequin . This analysis provides estimates of variance components and F-statistics [43] analogs reflecting the correlation of haplotype diversity at different levels of hierarchical subdivision . Unlike other approaches for partitioning genetic variation based on the analysis of variance of gene frequencies , AMOVA takes into account the genetic relatedness between molecular haplotypes . The hierarchical subdivision was made at three levels . At the top level , different groups were defined on the basis of the phylogenetic relationships for the different T . dimidiata haplotypes obtained . The second level corresponded to countries of sampling within each of these groups , and the third level corresponded to the different haplotypes found in each country within group . AMOVA reports components of variance at the three levels under consideration ( among groups , among countries within groups , and within countries within groups ) as well as F-statistics analogs . Under the present scheme , FST is viewed as the correlation of random haplotypes within countries within groups , relative to that of random pairs of haplotypes drawn from the whole species , FCT as the correlation of random haplotypes within groups , relative to that of random pairs of haplotypes drawn from the whole species , and FSC as the correlation of the molecular diversity of random haplotypes within countries within groups , relative to that of random pairs of haplotypes drawn from the corresponding group [44] . Although in the program used ( only currently available for molecular variance analysis ) the choice for establishing an intermediate level is fully arbitrary and has no influence on the final result of the comparison between units at the higher level , these same analyses were repeated by considering each haplotype , which may encompass several individuals , as a separate group for this intermediate level , because it could be argued that geopolitical country borders was not an appropriate choice despite its interest from the point of view of the control of Chagas disease . The statistical significance of fixation indices was tested using a non-parametric permutation approach [44] . Genetic differentiation between pairs of populations was evaluated by means of F-statistics [43] . Exact tests of population differentiation were performed [45] . Slatkin's linearized FST's [46] , [47] procedure was also followed to obtain estimates of pairwise equilibrium migration rates , both among groups , among countries within groups , and within countries for those cases in which haplotypes from more than one group were present .
The 137 ITS-2 sequences revealed the existence of 31 different haplotypes in the T . dimidiata studied ( T . dim-H1 to T . dim-H31 ) ( see Tables 1 and 2 for localities and countries ) . Their length was 489–497 base pairs ( bp ) ( mean , 495 . 10 ) with a relative AT-biased nucleotide composition of 75 . 25–76 . 85% ( 75 . 72% ) . Sequence similarity analysis of these 31 haplotypes revealed four distinct groupings: grouping 1 ( T . dim-H1 to T . dim-H10 ) ; grouping 2 ( T . dim-H11 to T . dim-H17 ) ; grouping 3 ( T . dim-H18 to T . dim-H24 ) ; and grouping 4 ( T . dim-H25 to T . dim-H31 ) ( Figure 2 ) . These four groupings appear linked to concrete wide geographical areas including neighboring countries and regions . The only exception is Providencia Island , which , although part of Colombia , is located 720 km off the northern coast of Colombia but only 240 km off the western coast of Nicaragua . No haplotype presents a very broad geographical distribution . The alignment of the 31 T . dimidiata haplotype sequences was 501 bp-long , of which 450 characters were constant and 24 were parsimony-informative . The interrupted microsatellite ( AT ) 4–5 TTT ( AT ) 5–7 was detected between positions 47 and 73 in all specimens studied . Variability in this microsatellite region and their respective sequence positions are noted in Figure 2 . The 51 nucleotide variable positions detected including gaps represented a 10 . 18% of polymorphic sites . The seven haplotypes T . dim-H25 to T . dim-H31 are responsible for this high genetic divergence ( Figure 2 ) . This genetic divergence decreases considerably when two separate alignments are performed: ( i ) the first includes T . dim-H1 to T . dim-H24 from all the seven countries shows a divergence of 5 . 62% in a 498-bp-long alignment , including 28 nucleotide variable positions , of which 6 ( 1 . 20% ) were transitions ( ti ) , 13 ( 2 . 61% ) transversions ( tv ) and 9 ( 1 . 81% ) insertions/deletions ( indels ) ; ( ii ) the second includes T . dim-H25 to T . dim-H31 from only three countries ( Mexico: localities of Yucatan , Chiapas , Cozumel Island and Holbox Island; Guatemala: Peten; Honduras: Yoro Yoro ) shows a divergence of 2 . 42% in a 495-bp-long alignment , with 12 nucleotide variable positions , of which 2 ti ( 0 . 40% ) and 10 are indels ( 2 . 02% ) . ITS-2 sequences of T . bassolsae , T . bolivari , T . hegneri , T . mexicana , T . pallidipennis , T . ryckmani , T . flavida , T . nitida , T . gerstaeckeri , and T . rubida , including haplotype length and AT content are listed in Table 1 . The comparison analyses which include these ITS-2 sequences and those of the Phyllosoma and Rubrofasciata groups ( available in GenBank ) provided 48 different haplotypes . Their alignment resulted in a total of 551 characters including gaps , of which 365 sites were constant and 99 parsimony-informative . All the T . dimidiata haplotypes clearly differed from the Phyllosoma , Flavida , Protacta and Rubrofasciata complex species included in this analysis . Triatoma bassolsae differed in only one deletion in position 489 from T . pallidipennis of Morelos , Mexico ( AJ286882 ) . The T . pallidipennis sequence obtained represents a new haplotype ( T . pal-H2 ) differing in only one deletion in position 31 from T . picturata and T . longipennis . The haplotype alignment of T . bassolsae , T . longipennis , T . mazzotti , T . picturata , T . pallidipennis and T . phyllosoma was 490 bp long showing a relatively small genetic diversity of 1 . 83% , with only 5 mutations ( 1 . 02% ) and 4 indels ( 0 . 81% ) . The two T . hegneri haplotypes differ between each other in only 1 ti and , when compared with T . dimidiata H18 to H24 from Mexico and Guatemala , nucleotide differences found were only 1 ti and 2 tv . ITS-2 sequences of T . maculata and T arthurneivai , including haplotype length and AT content are listed in Table 1 . The ITS-2 of T . maculata fits very well within sequences of the Infestans complex species studied in the present work , a total of 6–19 ( 13 . 7 ) mutations , namely 6–11 ( 7 . 25 ) ti and 0–10 ( 6 . 5 ) tv , appearing when comparing the five Infestans complex species in question . The material of Triatoma arthurneivai here analyzed is very close to T . rubrovaria H1 ( AJ557258 ) , showing only 6 nucleotide differences ( 1 . 22% ) , of which only 1 ti and 5 indels . Two different phylogenetic approaches were performed with the 31 T . dimidiata haplotypes , both yielding coincident results . A maximum likelihood tree was reconstructed using the best model of evolution as determined by the lowest AIC , which was GTR+I ( −Ln = 887 . 089 ) , being the proportion of invariable sites ( I ) of 0 . 166 . Three groups appeared with high support values indicating that their differentiation was not due to random sampling of a low variable sequence ( tree not shown ) . The large group 1 encompassed haplotypes from all the countries , whereas groups 2 ( Mexico and Guatemala ) and 3 ( Mexico , Guatemala and Honduras ) were more geographically restricted . Alternatively , a median-joining network was reconstructed with the 31 different T . dimidiata sequences using the variable sites in the multiple alignment ( Figure 3 ) . This network showed the same three groups found in the ML tree . Group 1 occupies a central position in the network and is the most widespread and variable group , so that it most likely corresponds to the ancestral or source set . This is further reinforced by the direct relationship between this group and the two others , more geographically restricted and encompassing fewer variants , group 2 including samples from Mexico and Guatemala , and group 3 including samples from these two countries and Honduras . The group 1 source set would in turn be derived from group 3 , which might be interpreted as a geographically restricted relict according to the phylogeographic results . Moreover , sequence variants in group 1 are clustered in two different subgroups , with genetic and geographical borders: subgroup 1A includes sequences from Colombian Providencia island , Ecuador , Guatemala , Honduras , Mexico ( only South of Chiapas ) and Nicaragua; subgroup 1B encompasses sequences from continental Colombia and Panama . The two closest sequences of each subgroup differ in two sites , which might correspond to haplotypes not found in this sampling . The relevance of the ITS-2 differences among these T . dimidiata groups and subgroups was assessed by comparison with other Triatoma species . Therefore , a multiple , 562-nucleotide-long alignment was obtained by incorporating 22 additional ITS-2 sequences . This set includes 53 ITS-2 sequences of Triatoma species and , using R . prolixus as outgroup , a ML tree was obtained ( −Ln = 2648 . 5129 ) using the HKY+G model , according to the AIC results with a gamma distribution shape parameter = 0 . 58 . This tree ( Figure 4 ) shows that: Triatoma dimidiata groupings appeared well supported , with very high bootstrap proportions ( BP>90% ) using ML and neighbor-joining reconstruction and the highest Bayesian posterior probabilities ( BPP = 100% ) . Similar levels were found for other well established Triatoma species , many of which showed substantially lower support values in the three statistical measurements employed . However , other species presented no ITS-2 nucleotide differences ( T . picturata and T . longipennis; T . mazzotti and T . phyllosoma ) . The phylogenetic analyses showed that samples from the same country may belong to different clusters . This result , on its own , is not enough to demonstrate the biological distinctiveness of the corresponding populations . Sampled individuals may represent a minor fraction of the total genetic variability in a highly heterogeneous population and the sampling procedure might have resulted , by pure chance , in the observed clustering of some variants . Given that each of these clusters holds some genetic variability of its own , the first task was to evaluate whether the observed groupings were significantly different from each other , in terms of genetic variation , by partitioning the observed genetic variability at three different levels: among groups , among populations ( countries ) within groups , and within populations . A hierarchical analysis of molecular variance was used to test the null hypothesis of no genetic differentiation among groups considering variation at lower levels . This procedure was first applied to T . dimidiata sequences using three levels as defined above ( Table 3a ) . Most of the genetic variation found was allocated to the among groups level ( 80 . 24% of the total variation ) , with much lower portions of variation assigned to differences among populations within groups level ( 11 . 71% ) and within populations level ( 8 . 05% ) , although both were still statistically significant after 1000 pseudo-random samples generated for testing . This indicates that , despite genetic variation within and among populations at these three levels , there is a substantial amount of genetic differentiation among them that justifies their consideration as separate groupings for further analysis . The same results were obtained , notwithstanding small numerical differences due to the different numbers of groups , when haplotypes instead of countries were considered at the intermediate level ( Table S1 ) . The geographical fitting represents in fact no surprise at all , taking into account that the distribution of T . dimidiata covers different countries which are more or less aligned following a north-south axis because of the relatively slenderness of the Central American bridge . Hence , as any of the two versions of the analyses conveys the same information and leads to the same conclusions , and which one should be reported is simply a matter of opinion , the first considering countries becomes practically more useful because Chagas disease control measures are organized at national level . The median-joining network reconstructed with the 31 different T . dimidiata ITS-2 sequences revealed the existence of three distinct groups ( groups 1 , 2 and 3 ) , the first of which further subdivided into two subgroups 1A and 1B . The same AMOVA procedure was applied to ascertain whether these two subgroups could be considered as distinct populations or not . The results ( Table 3b ) indicate that a significant fraction ( 60 . 15% ) of the total genetic variation corresponds to differences between these two subgroups which , correspondingly , could be considered as separate populations for the ensuing analyses . Based on the four groups/subgroups previously described in the median-joining network , a summary of relevant population genetic parameters for T . dimidiata is presented in Table 4 . Genetic variation in T . dimidiata populations was quite evenly distributed , with similar levels of nucleotide and haplotype diversities in the four groups/subgroups considered . Nevertheless , for all the parameters studied , subgroup 1A presented higher values than the rest , although significance of the differences was only obtained for haplotype diversity . A similar summary is shown for each country sample within groups in Table S2 . Different estimates of θ were obtained based on the expected heterozygosity , the expected number of alleles , the number of polymorphic sites and the nucleotide diversity . The four estimates were quite consistent for the four groups/subgroups and they agreed in assigning a larger value to subgroup 1A . Differences in the genetic composition of the four groups/subgroups 1A , 1B , 2 and 3 have previously been shown to be statistically significant according to analyses of molecular variance . A further evaluation of this distinctiveness was made ( Table 3c ) , in which the four groups/subgroups were considered for the AMOVA , in correspondence with the previous results . In this case , the amount of among-group variation rose to 86 . 84% of the total variation , whereas among population within groups and within population levels they were substantially lower , 3 . 21% and 9 . 95% respectively . Genetic differences within and among the ITS-2 locus for T . dimidiata samples were further explored through pairwise comparisons , and estimates of average pairwise differences within and among the four groups/subgroups considered were obtained ( Table 5 ) . Subgroup 1A presented the largest value for within-group pairwise differences . The within-population values were much lower than among-populations comparisons . Among the latter , the smallest number of differences was found between subgroup 1A and 1B , in correspondence with their close phylogenetic relationship . Subgroup 1B was the one with the lowest overall number of pairwise differences , slightly below 1A . On the contrary , the highest value of pairwise differentiation corresponds to group 3 , with almost 20 differences ( corrected estimate ) when compared with any other group . Within groups genetic differentiation was evaluated by computation of pairwise FST values for populations defined by country of origin ( Table S3 ) . Since all groups/subgroups , with the only exception of subgroup 1A , are characterized by one large ( n>10 ) and several small ( n<10 ) populations , significance values for test of genetic differentiation have to be interpreted cautiously . Hence , there is no apparent differentiation between two populations in subgroup 1B ( Colombia2 , n = 30 , and Panama , n = 4 ) and similarly in group 2 ( Mexico2 , n = 23 , and Guatemala2 , n = 4 ) . The only significant value found in group 3 corresponds to Honduras3 ( n = 2 ) and Guatemala3 ( n = 7 ) , for which FST = 0 . 529 , P<0 . 05 . None of these two populations presented significant differentiation with respect to the largest population in this group , Mexico3 ( n = 15 ) . Subgroup 1A includes two large populations , Honduras1 ( n = 18 ) and Guatemala1 ( n = 26 ) , which presented a highly significant FST = 0 . 193 , P<0 . 001 . Although this value , under the assumption of migration-drift equilibrium , corresponds to an estimate of 2 . 1 migrants per generation between both populations , which would be enough to prevent their complete differentiation , such estimations shall be verified by using larger samples and markers better suited for population genetics analyses . Comparisons between each of these two populations and the smaller ones in subgroup 1A revealed that Honduras1 differed from Mexico1 , Guatemala1 was different from Ecuador and Nicaragua , and none of them differed from the only two individuals from Providencia island . Similar comparisons for all pairs of populations assigned to different groups/subgroups resulted in highly significant FST values ( Table S4 ) .
The highest intraspecific ITS-2 variability ( absolute nucleotide differences including indels ) known in Triatomini members is 2 . 70% ( 13/482 ) in T . infestans specimens collected throughout the very wide geographical distribution of this species [25] . Hence , the result of 10 . 18% ( = 51/501 ) detected in T . dimidiata ( Figure 2 ) appears to be pronouncedly outside the limits of the intraspecific variability range known for Triatoma species . Group 3 is the main responsible for such differences ( Table 5 ) and shows a high 2 . 42% divergence within itself , suggesting an old origin in the light of the relatively reduced geographical area of distribution of these haplotypes in Mexico ( Yucatan , Chiapas , Cozumel Island and Holbox Island ) , Guatemala ( Peten ) and Honduras ( Yoro ) only . The time of divergence between group 3 and other T . dimidiata populations was estimated to be of 5 . 9–10 . 5 million years ago ( Mya ) according to a molecular clock analysis based on rDNA evolutionary rates [4] . The divergence of 5 . 62% shown by the other 24 ITS-2 haplotypes ( Figure 2 ) also appears to be too large , in spite of the wide geographical area they occupy from Mexico down to Ecuador , suggesting a speciation process . However , population average pairwise differences between subgroup 1A , subgroup 1B and group 2 are markedly lower than between these three and group 3 ( Table 5 ) , and intragroup differences do fall within the above-mentioned Triatomini range: 2 . 61% within subgroup 1A , 2 . 41% within subgroup 1B , and 2 . 01% within group 2 . Results indicate that several T . dimidiata populations are following different evolutionary divergences in which geographical isolation appears to have had an important influence ( Figure 5 ) . A phenotypic consequence of that process had been observed by other specialists before , who wrote about an assemblage of morphologically variable populations [10] . More recently , significant head shape differences between populations showed a separation between northern , intermediate and southern collections of T . dimidiata and also support an evolutionary divergence of populations within this species [13] . Three subspecies were distinguished on the basis of morphological differences [48] , [49]: ( i ) T . d . dimidiata concerns the first description of the species in Peru ( no type specimen available; no type locality assigned , but undoubtedly from northern Peru , probably around the locality of Tumbes , near Ecuador ) and corresponds to most of the Central American forms; ( ii ) T . d . maculipennis was proposed for specimens from Mexico ( type specimen in Zoologisches Museum Berlin ) and corresponds to forms with relatively short heads and large eyes; and ( iii ) T . d . capitata was proposed for large size specimens typified by longer heads and smaller eyes originally found in Colombia ( type specimen in the Academy of Sciences of California ) . However , these subspecies became later synonymized after results of a morphological re-examination which were interpreted as evidence of a clinal variation along a north-south axis [50] , [51] . Present ITS-2 sequences and corresponding phylogenetic and genetic variation analyses support the appropriateness to ( i ) differentiate group 3 as a species of its own ( here simply designed as T . sp . aff . dimidiata to avoid further systematic confusion with T . dimidiata , according to taxonomic rules ) , and ( ii ) re-assign subspecific status for subgroup 1A , subgroup 1B and group 2 . Results of the present study do not support the rise of the above-mentioned subspecific taxa to species level for the time being , although it is evident that in the three cases relatively long divergence processes have taken place . Similar genetic studies with other molecular markers may contribute to a more complete assessment of these evolutionary isolation and speciation processes . The taxon T . sp . aff . dimidiata concerns group 3 . This species seems to represent a relatively relict species with a distribution restricted to the Mexican flat areas of the Yucatan peninsula and the northern part of Chiapas state , the northern lowland of Guatemala ( and probably also Belize ) , and only one altitude-adapted haplotype ( T . dim-H29 ) in its most extreme border populations in northern Honduras . The most widely spread haplotype T . dim-H28 is also present in the small island of Holbox and the large island of Cozumel , both near the Yucatan coast , suggesting that this haplotype should be considered the oldest of this species . This species is also of public health importance because of its capacity to transmit Chagas disease [52] , [53] and the control problems it poses [54] , [55] . The taxon T . d . dimidiata corresponds to subgroup 1A and populations mainly from Guatemala and Honduras and secondarily Mexico , Nicaragua and Ecuador . The population of the Colombian island of Providence undoubtedly derives from the most widely dispersed haplotype T . dim-H1 on the nearest Caribbean coastal area of Central America and not from continental Colombia . The present populations in Ecuador may derive from introduced specimens originally from the Guatemala-Honduras-Nicaragua region , relatively recently introduced by humans [4] , very probably in the period of the early colonialization of northwestern South America by the Spanish ‘conquistadores’ in which exchange activities between Central American settlements and the Peruvian Tumbes area took place [56] . The type specimens of the original description of the species in northern Peru might also belong to populations derived from such man-made introductions from Central America . The haplotype T . dim-H10 of Lanquin , Alta Verapaz , Guatemala appears in the network analysis as directly derived from an ancestor which gave rise to the subspecies T . d . dimidiata . An isolation phenomenon in caves may explain the albinic characteristics of the specimens presenting this haplotype . These cavernicole specimens from Alta Verapaz have already shown their peculiarity in morphometric and cuticular hydrocarbon studies [13] , [17] . The taxon T . d . capitata corresponds to subgroup 1B and populations from Colombia and Panama . The isthmus of Panama and the separation/joining process of South and North America towards the end of the Pliocene ( 3–5 Mya ) [57] , in a period in which several more or less closely separated islands appeared and evolved up to their fusion into the isthmus , should have played a major role in the isolation and subsequent divergence of these southernmost T . dimidiata populations . The lack of relationship between the haplotypes of Ecuador and those of Colombia is worth mentioning , as the geographical closeness of these two countries could have given rise to the erroneous hypothesis of Colombian forms having derived from Ecuadorian populations . In a recent study of three populations of sylvatic , peridomestic and domestic T . dimidiata from Colombia , the estimated low genetic distances based on RAPD analyses did not discriminate the populations studied , indicating that they maintain the genetic identity of a single recent common ancestor [9] . The taxon T . d . maculipennis corresponds to group 2 and populations mainly from Mexico , but rarely found in Guatemala . According to the network analysis , this subspecies seems to have derived from group 1 probably by isolation in the Mexican part northward from the isthmus of Tehuantepec . Similarly as for other organisms including insects [58] , the mountainous Sierra Madre chain throughout southern Mexico and Guatemala areas near the Pacific coast probably played also a role in that isolation process through an area where T . sp . aff . dimidiata did not represent a competition barrier , as T . sp . aff . dimidiata appears to be preferentially a low altitude species in these two countries . Southern Mexico ( including the Yucatan peninsula and Chiapas state ) and almost the whole country of Guatemala ( at least ten departments ) constitute a crucial evolutionary area , where a high number of taxa , including T . d . dimidiata , T . d . maculipennis , and T . sp . aff . dimidiata , overlap . In a morphometric analysis , populations from San Luis Potosi and Veracruz in Mexico were indistinguishable while clearly different from populations from Yucatan in Mexico and Peten in Guatemala [14] . The former correspond to T . d . maculipennis and the latter to T . sp . aff . dimidiata . In Guatemala , a high degree of genetic variation in T . dimidiata sensu lato was shown by RAPD-PCR [12] , demonstrating a limited gene flow between different provinces , although barriers between the Atlantic and Pacific drainage slopes did not appear to be significant limiters of a gene flow , according to a hierarchical analysis . Chromosome analyses and DNA genome size revealed the existence of three different cytotypes with different geographical distributions [18]: ( i ) cytotype 1 corresponds to three different taxa: T . d . maculipennis in Mexico ( excluding Yucatán ) , T . d . dimidiata in Guatemala ( excluding Petén ) and probably also El Salvador; and T . d . capitata in Colombia; ( ii ) cytotype 2 was found in two localities ( Paraiso and Chablekal ) around Mérida , Yucatan , Mexico where the species T . sp . aff . dimidiata presents 5 different haplotypes ( T . dim-H25 , T . dim-H26 , T . dim-H27 , T . dim-H28 and T . dim-H31 ) ; ( iii ) cytotype 3 appeared in Yaxhá , Petén , Guatemala , where both T . d . maculipennis ( T . dim-H18 ) and T . sp . aff . dimidiata ( T . dim-H25 , T . dim-H28 and T . dim-H30 ) are present . Sequencing of the same specimens studied [18] from Yaxhá showed that cytotype 3 was found in specimens of T . sp . aff . dimidiata of haplotype T . dim-H28 and T . dim-H30 . Consequently , chromosomal cytotypes 2 and 3 are both found in T . sp . aff . dimidiata . The two haplotypes of T . hegneri differ by only 3 mutations from haplotypes of T . d . maculipennis . This reduced number of nucleotide differences and the location of T . hegneri haplotypes within the clade of T . dimidiata , basal to haplotypes of group 2 ( Figure 4 ) , does not support its status as an independent species . The results obtained suggest that it is an insular form of T . d . maculipennis . Originally described from the island of Cozumel [3] , a subspecific status T . d . hegneri could be maintained only if morphological characteristics allow a clear differentiation of the insular form , as the phylogenetic analysis somehow separates it in a very close but particular evolutionary line . Triatoma hegneri , although chromatically distinguishable from most forms of T . dimidiata [50] , is known to produce fertile hybrids when experimentally crossed with T . dimidiata ( R . E . Ryckman , unpublished ) . Interestingly , the most dispersed haplotypes of both T . d . maculipennis ( T . dim-H18 ) and T . sp . aff . dimidiata ( T . dim-H28 ) are also present on the same island , probably introduced through the intense human transport between the mainland and the island . The distinction between T . d . dimidiata ( subgroup 1A ) , T . d . capitata ( subgroup 1B ) , T . d . maculipennis ( group 2 ) , T . sp . aff . dimidiata ( group 3 ) , and T . d . hegneri contributes giving systematic/taxonomic coherency to present knowledge about morphological and genetic concepts in these taxa . From an ancestral form close to T . sp . aff . dimidiata , it can be postulated that an original diversification focus of T . dimidiata forms took place most probably in Guatemala , with a southern spread into Panama and Colombia to give the capitata forms and a northwestern spread into Mexico to give the maculipennis forms ( Figure 5 ) . Thus , the results of the present paper , obtained from a large amount of samples of T . dimidiata from many different countries covering its whole latitude range , gives rise to a new frame that is different from the previous hypothesis about a clinal variation along a north-south axis , which was formerly suggested to explain both morphological data [50] and preliminary ITS-2 data from a reduced number of samples [6] . Moreover , the distinction between these five entities may facilitate the understanding of different vector transmission capacities and epidemiological characteristics of Chagas disease throughout the very large area where T . dimidiata sensu lato is distributed , from the Mexican northern latitude limit up to the Peruvian southern latitude limit [11] . Recent results obtained by means of a population dynamics model indicate that T . dimidiata in Yucatan , Mexico , is not able to sustain domestic populations , that up to 90% of the individuals found in houses are immigrants , and that consequently Chagas disease control strategies must be adapted to a transmission by non-domiciliated vectors [59] . This might be considered surprising because it does not fit the domiciliation capacity of T . dimidiata in other places , but it appears to be congruent if it is taken into account that in fact the Yucatan vector in question is not T . dimidiata but a different species T . sp . aff . dimidiata . The results here obtained also suggest that T . d . dimidiata in Ecuador is a good candidate for the design of appropriate vector control intervention , similarly to domestic T . infestans populations in countries such as Uruguay , Chile and Brazil within the successful Southern Cone Initiative [60] . The control and even eradication of T . d . dimidiata in Ecuador by means of insecticide-spraying of its domestic habitats might be successful , if it is considered that it is merely an introduced vector species in that area , and a priori it would have difficulties in escaping from the insecticide activity because of its non-adaptativeness to the sylvatic environment in these two countries [61] . Unfortunately , such a control initiative will not be so easy to carry out in Colombia , as results prove that Colombian forms are authochthonous T . d . capitata and not T . d . dimidiata derived from the Ecuadorian introduced form . This fits with the existence of sylvatic populations in Colombia and with the high genetic similarity of sylvatic , peridomestic and domestic populations detected in that country [9] . Similarly to in Colombia , results indicate that T . dimidiata will offer , because of being authochthonous forms , more problems for insecticide-spraying control in Central American countries than introduced T . infestans in Southern Cone countries . Triatoma bassolsae differs by only one deletion from T . pallidipennis and appears in the branch of the 5 species traditionally included in the Phyllosoma complex: T . longipennis , T . mazzotti , T . picturata , T . pallidipennis and T . phyllosoma . The genetic differences between these taxa are so reduced ( sometimes even none at all ) , that there is no support to maintain them as separated species . Such a low number of nucleotide differences in the ITS is considered as pertaining to organisms able to hybridize [62] . This fully fits the capacity of these taxa to crossbreed and give fertile hybrids [63] , [64] and agrees with the entomologist conclusion of applying only subspecies level to them [49] . The divergence of members of the phyllosoma complex is estimated at only 0 . 74–2 . 28 Mya by the rDNA molecular clock [4] , which also seems consistent with a subspecific rank . All further ITS-2 studies have always reached the same conclusion [5] , [6] , [65] . By analyzing many interfertility experiments [64] , it can be concluded that , in triatomines , morphological differentiation appears to be faster than the installation of reproductive or genetic barriers [66] , [67] . Rapid morphological changes , associated with ecological adaptation , helps to explain discordance between phenetic and genetic differentiation . Triatomine species with consistent morphological differences would arise through divergent ecological adaptation , a vision which fits with “evolutionary units” implying a different evolutionary direction taken by some populations [67] . Until future reproductive isolation thanks to ecological isolation is reached by these morphologically different entities of the Phyllosoma complex , the subspecies concept accurately fits for all these “evolutionary units” of the Phyllosoma complex . ITS-2 results indicate that Triatoma bassolsae is one additional taxon to be included in this situation , as has already been suggested [65] . The comparison of the small genetic divergences between these taxa , their distributions exclusively restricted to regions of Mexico , and their different geographical distribution areas slightly overlapping in their bordering zones [3] suggest that genetic exchange might be impeding or delaying definitive divergence processes to reach species level . Genetic distances between the taxa of the Phyllosoma complex found when analyzing different mtDNA genes proved to be similar to those detected in ITS-2 at the 16S [68] , but higher in CytB [65] , [69] , and COI [69] . This agrees with the evolutionary rates of the protein-coding mtDNA genes which are pronouncedly faster than the one of ITS-2 . Moreover , aminoacid sequences of the CytB and COI genes show no one difference between the Phyllosoma complex members studied ( all are silent mutations or synonymous substitutions ) except one aminoacid difference between two populations of the same species T . pallidipennis and one in T . picturata versus the rest [69] , which also fit with an intraspecific variability . Additionally , it shall be taken into account that ( i ) mtDNA becomes monophyletic more rapidly than does a single nuclear gene and far more rapidly than a sample of several nuclear genes , so that mtDNA may make inferences of species-level monophyly erroneous [70] , and ( ii ) the known great potential of mtDNA to become monophyletic by selective sweeps can decrease the time to monophyly of a clade and not be reflective of the genealogical processes in the nuclear genome , advantageous mutations occurring on mtDNA causing the entire mitochondrial genome to become monophyletic because of the little or no recombination they have [71] . The crossbreeding capacity and hybrid viability among the Phyllosoma complex taxa in question is well known and , taking into account that their geographical distributions overlap in their border areas and there are no sufficient ecological differences indicating a local spatial separation , it becomes very difficult to support them as separate species from the evolutionary , biogeographical and ecological points of view because there is apparently no barrier for a reproductive isolation . Thus , the results of both ITS-2 and mtDNA genes fit with such an evolutionary , subspecific divergence , when taking into account the peculiarities of both nuclear and mitochondrial markers . Triatoma mexicana appears to be a good species and its location in the phylogenetic tree fully supports its ascription to the Phyllosoma complex , similarly as suggested by a phylogentic analysis by means of a mtDNA CO1 fragment [69] . Surprisingly , T . gerstaeckeri ( Rubrofasciata group ) clusters with T . mexicana , suggesting that it should be included in the Phyllosoma complex . All these species , i . e . T . phyllosoma ( including its subspecies phyllosoma , longipennis , mazzotti , picturata , pallidipennis and bassolsae ) , T . dimidiata ( with its three subspecies dimidiata , capitata and maculipennis , to which hegneri shall be added ) , T . sp . aff . dimidiata , T . mexicana and T . gerstaeckeri constitute a well defined clade for which the generic taxon Meccus , proposed long ago [72] , afterwards synonymized [50] and recently tentatively revalidated [73] , seem to appropriately fit . Previous molecular studies , first with complete ITS-2 sequences [74] and second with partial mtDNA 16S gene sequences [68] , also indicate that Meccus might be a valid taxon . The revalidation of Meccus , as well as that of Nesotriatoma for species of the Flavida complex , has not been accepted because of the close relationship between T . flavida and the Phyllosoma complex [7] . The results of the present study do , however , pose a serious question concerning the inclusion of species as T . bolivari and T . ryckmani in the Phyllosoma complex , as they appear to cluster with T . rubida of the Rubrofasciata group with relatively high support ( 83 and 96 in ML and BPP , respectively ) . A T . rubida - T . nitida clade previously detected with weak support under certain conditions in mitochondrial DNA marker analyses [69] does not appear to be supported in the ITS-2 phylogeny . Although not fully resolved in the tree obtained , the location of the Cuban T . flavida as a species basal to all other North-Central American Triatoma species may be interpreted as a consequence of being a relict insular species close to the ancient first North-Central American Triatoma colonizers . Further studies with other genetic markers are needed to establish the position of T . flavida more adequately . The very scarce ITS-2 sequence differences between T . arthurneivai and T . rubrovaria , a species known in southern Brazil , Uruguay and northern Argentina [75] , pose doubts on whether to keep the validity of T . arthurneivai as independent species . Recent genetic and morphometric studies have already raised several questions about T . arthurneivai , indicating that topotypes from Minas Geraes may represent a species different from populations of São Paulo State formerly also referred to T . arthurneivai and suggesting that these São Paulo populations might probably belong to T . wygodzinskyi [76] . This may explain the ITS-2 results , as the two specimens analyzed in the present paper come in fact from Espirito Santo do Pinhal , São Paulo State . Consequently , material of typical T . wygodzinskyi should be sequenced and compared to both true T . arthurneivai from Minas Geraes and T . rubrovaria to ascertain the status of these three taxa . The South American Triatoma species cluster together with maximum support ( 100/100/100 ) and well separated from that of the North and Central American species of the same genus , thus supporting results of previous analyses which indicate an early divergence of about 23–38 Mya between species of the northern ( Phyllosoma complex ) and southern ( T . infestans ) continent [4] , [6] . | Chagas disease is a serious parasitic disease of Latin America . Human contamination in poor rural or periurban areas is mainly attributed to haematophagous triatomine insects . Triatoma includes important vector species , as T . dimidiata in Central and Meso-America . DNA sequences , phylogenetic methods and genetic variation analyses are combined in a large interpopulational approach to investigate T . dimidiata and its closest relatives within Triatoma . The phylogeography of Triatoma indicates two colonization lineages northward and southward of the Panama isthmus during ancient periods , with T . dimidiata presenting a large genetic variability related to evolutionary divergences from a Mexican-Guatemalan origin . One clade remained confined to Yucatan , Chiapas , Guatemala and Honduras , with extant descendants deserving species status: T . sp . aff . dimidiata . The second clade gave rise to four subspecies: T . d . dimidiata in Guatemala and Mexico ( Chiapas ) up to Honduras , Nicaragua , Providencia island , and introduced into Ecuador; T . d . capitata in Panama and Colombia; T . d . maculipennis in Mexico and Guatemala; and T . d . hegneri in Cozumel island . This taxa distinction may facilitate the understanding of the diversity of vectors formerly included under T . dimidiata , their different transmission capacities and the disease epidemiology . Triatoma dimidiata will offer more problems for control than T . infestans in Uruguay , Chile and Brazil , although populations in Ecuador are appropriate targets for insecticide-spraying . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"evolutionary",
"biology/animal",
"genetics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"molecular",
"biology/molecular",
"evolution",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2008 | Phylogeography and Genetic Variation of Triatoma dimidiata, the Main Chagas Disease Vector in Central America, and Its Position within the Genus Triatoma |
Recent studies have shown that exposure to some nutritional supplements and chemicals in utero can affect the epigenome of the developing mouse embryo , resulting in adult disease . Our hypothesis is that epigenetics is also involved in the gestational programming of adult phenotype by alcohol . We have developed a model of gestational ethanol exposure in the mouse based on maternal ad libitum ingestion of 10% ( v/v ) ethanol between gestational days 0 . 5–8 . 5 and observed changes in the expression of an epigenetically-sensitive allele , Agouti viable yellow ( Avy ) , in the offspring . We found that exposure to ethanol increases the probability of transcriptional silencing at this locus , resulting in more mice with an agouti-colored coat . As expected , transcriptional silencing correlated with hypermethylation at Avy . This demonstrates , for the first time , that ethanol can affect adult phenotype by altering the epigenotype of the early embryo . Interestingly , we also detected postnatal growth restriction and craniofacial dysmorphology reminiscent of fetal alcohol syndrome , in congenic a/a siblings of the Avy mice . These findings suggest that moderate ethanol exposure in utero is capable of inducing changes in the expression of genes other than Avy , a conclusion supported by our genome-wide analysis of gene expression in these mice . In addition , offspring of female mice given free access to 10% ( v/v ) ethanol for four days per week for ten weeks prior to conception also showed increased transcriptional silencing of the Avy allele . Our work raises the possibility of a role for epigenetics in the etiology of fetal alcohol spectrum disorders , and it provides a mouse model that will be a useful resource in the continued efforts to understand the consequences of gestational alcohol exposure at the molecular level .
While it is well-recognized that gestational exposure to environmental triggers can lead to compromised fetal development and adult disease in humans [1] , the underlying molecular mechanisms remain unknown . There is increasing evidence in animal models that environmental factors can affect gene expression via epigenetic modifications such as DNA methylation [2]–[6] . One way of detecting such events is to use reporters whose expression is closely linked to their epigenetic state . Such epigenetically sensitive alleles are also known as metastable epialleles , and the best known example in the mouse is Agouti viable yellow ( MGI:1855930 ) or Avy [7] . Avy is a dominant mutation of the murine Agouti ( A ) locus , caused by the insertion of an intracisternal A-particle ( IAP ) retrotransposon upstream of the Agouti coding exons . The activity of Avy is variable among genetically identical mice , resulting in mice with a range of coat colors; from yellow to mottled to agouti ( termed pseudoagouti ) [8] . The expression of Avy is known to correlate with DNA methylation at a cryptic long terminal repeat ( LTR ) promoter located at the 3′ end of the inserted IAP . Specifically , hypomethylation is associated with constitutive ectopic Agouti expression and a yellow coat , while hypermethylation correlates with cryptic promoter silencing and a pseudoagouti coat [9] . We have previously shown that DNA methylation at Avy is reprogrammed in early development at the same time that the rest of the genome is undergoing epigenetic reprogramming [10] . Alcohol consumption is widespread in our society , but it is also recognized as the leading preventable cause of birth defects and mental retardation [11] , [12] . High levels of alcohol consumption during pregnancy can result in fetal alcohol syndrome ( FAS ) which is characterized by prenatal and postnatal growth restriction , craniofacial dysmorphology and structural abnormalities of the central nervous system . The clinical features of FAS are variable and include a range of other birth defects , as well as educational and behavioral problems [13] . This syndrome is the most extreme form of a range of disorders that are known as fetal alcohol spectrum disorders ( FASDs ) [14] . Approximately 5% of the children of mothers who have drunk heavily during pregnancy have FAS [15] , and studies have shown that the dose , time and duration of ethanol exposure are critical [16] , [17] . There are a number of mouse models of FAS that have reproduced some of the phenotypic characteristics of the human disorder , particularly the craniofacial abnormalities [16] , [18] , [19] . It should be noted that these studies used acute ethanol exposures between gestational days ( GDs ) 7 and 9 and high concentrations; generally two intraperitoneal injections of 0 . 015 ml of ∼25% ( v/v ) ethanol per gram of body weight over a 4 hour interval resulting in ataxia and lethargy . These studies only examined the fetal outcomes ( GDs 8-18 ) of ethanol exposure and did not assay offspring either after birth or as adults . There are some rodent studies of the effects of gestational exposure to moderate amounts of ethanol , but these have only identified neurological and behavioral deficits [20] . The molecular mechanisms underlying FAS are unknown . Some studies have focused on the toxic effects of acetaldehyde , the first metabolite of ethanol [18] , [21] . Acute ethanol exposure has also been found to result in increased cell death in the developing central nervous system and neurological anomalies in rodents and other animal models [22] , [23] . The idea that epigenetic changes are involved has been raised but evidence in support of this hypothesis has , so far , been weak . Garro and colleagues [24] detected a small decrease in the level of global methylation of fetal DNA after acute ethanol administration from GDs 9-11 . Bielawski et al . [25] reported decreased DNA methyltransferase 1 ( Dnmt1 ) messenger RNA levels in rat sperm after nine weeks of paternal ethanol exposure . Haycock and Ramsey [26] studied imprinting of the H19/Igf2 in preimplantation mouse embryos after maternal ethanol exposure . Despite severe growth retardation of embryos , they did not find epigenetic changes at the H19 imprinting control region . Here we have developed a mouse model of chronic ethanol exposure ( overt signs of intoxication are not observed ) that produces measurable phenotypes in adults . We find that maternal ethanol consumption either before or after fertilization affects the expression of an epigenetically sensitive allele , Avy , in her offspring and that , at least in the latter case , can also impact postnatal body weight and skull size and shape in a manner consistent with FASD . Our work raises the possibility of a role for epigenetic reprogramming in the etiology of FASD and provides researchers with a relevant mouse model of the human disorder .
In this study , Avy was used primarily as a sensitive reporter of epigenetic changes in response to maternal ethanol consumption . The C57BL/6J mouse is null ( a ) at the Agouti locus , so it has a black coat color . Avy is a gain-of-function , semi-dominant mutation and so the coat color of heterozygous ( Avy/a ) mice in the C57BL/6J background is a direct read out of Avy transcriptional activity and DNA methylation . The nature of the matings used in this study , an Avy/a male crossed with an a/a female , means that only 50% of the offspring will inherit the Avy allele and be useful for coat color phenotyping . The remaining ( a/a ) offspring will be black . To study the effects of gestational ethanol exposure , female a/a C57BL/6J mice were supplied with 10% ( v/v ) ethanol in their drink bottles for eight days after fertilization by a congenic male carrying the Avy allele ( n = 46 litters , 242 total offspring , 109 Avy/a offspring ) . To evaluate the effects of preconceptional ethanol exposure , female a/a mice were given 10% ( v/v ) ethanol for four days per week for ten weeks prior to fertilization ( n = 22 litters , 131 total offspring , 69 Avy/a offspring ) . The Avy allele was passed through the male germ line to avoid the bias associated with maternal transmission , where epigenetic marks can be incompletely cleared between generations [9] . Control mice were given water instead of ethanol ( n = 37 litters , 189 total offspring , 91 Avy/a offspring ) . Maternal ethanol exposure during gestation did not significantly alter Mendelian inheritance of the Avy allele ( data not shown ) or litter size ( control 5 . 1±0 . 4 , ethanol exposed 5 . 2±0 . 3 , mean±SEM , Student's t-test , p = 0 . 9 ) . The establishment of epigenetic marks at Avy occurs during early embryogenesis and is a probabilistic event . The resulting variable expression of Avy among genetically identical mice produces individuals with a predictable range of coat colors . We found that , in the absence of any treatment , 21% of the offspring of Avy/a sires were yellow , 66% were mottled and 13% were pseudoagouti ( Figure 1 ) . Gestational ethanol exposure resulted in a higher proportion of pseudoagouti ( Pearson's chi-square test , p<0 . 05 ) . Twenty-eight percent of offspring were pseudoagouti compared with 13% in the control group ( Figure 1 ) . Preconceptional ethanol exposure produced a similar trend ( Pearson's chi-square test , p<0 . 05 ) . This shows that ethanol exposure can influence the establishment of Avy expression early in development . It increases the probability of transcriptional silencing at this particular locus . To confirm that the coat color correlated with DNA methylation at the Avy allele in gestationally exposed mice , 11 CpG dinucleotides in the LTR cryptic promoter of the Avy IAP were subjected to bisulfite sequencing ( Figure 2 ) . The results showed that , as expected , ethanol-exposed yellow mice were hypomethylated compared to ethanol-exposed pseudoagouti mice . Interestingly , atypical hypermethylated clones were found in five out of six yellow mice in the ethanol-exposed group , but they were clearly not sufficient to affect coat color . In the ethanol-exposed group 11% of the CpG dinucleotides were methylated compared to 2% in the control group . Using this measure a Student's t-test or non-parametric equivalent was unsuitable because the data did not meet the distribution requirements of being spread on a continuum . So we analyzed allele-specific methylation . In the ethanol-exposed group 23% of clones showed evidence of methylation , n = 91 , compared with 8% of clones in the control group , n = 71 ( Pearson's chi-square tests , p<0 . 01 ) . In contrast , total DNA methylation level in ethanol exposed pseudoagouti mice ( 61% ) was not significantly different to that observed in the controls ( 65% , Student's t-test , p = 0 . 27 ) . Equivalent results were obtained from a random effects model which allowed for the clustering of clones within mice ( p = 0 . 23 ) . Bisulfite sequencing was also carried out on control and ethanol-exposed mottled mice and we found the results extremely variable from one mouse to the next within both groups ( Figure S1 ) . Presumably , this is the result of the small size of the tissue sample ( tail tip ) . The variegated expression in mottled mice means that any one sample could represent only one clonal patch , which could harbour an active or an inactive Avy allele and not represent the true methylation state of the whole animal . For this reason mottled mice were not used in our analyses . The effects of Avy expression are pleiotropic . For example , yellow mice exhibit hyperphagia , hyperglycemia , non-insulin-dependent diabetes and adult onset obesity [27] . We did not assay these other phenotypes following ethanol exposure in our mice as their relevance to humans is questionable since no human ortholog of the Avy allele has been identified . So , while Avy was initially useful as a sensitive indicator of epigenetic changes , any further study of FAS-like phenotypes must necessarily focus elsewhere in the genome . For this reason and the fact that variable Agouti expression would confound many phenotypes , all subsequent analyses were performed on the congenic a/a siblings of the Avy mice . In the mouse , IAPs are present at approximately 1 , 000 copies per haploid genome [28] . To see if gestational ethanol exposure changed the methylation level of IAPs globally , we performed bisulfite sequencing using PCR primers that anneal to all IAP LTRs and analyzed ten CpG sites in both the tail and forebrains of a/a mice . Tail DNA from eight ethanol-exposed mice ( 66 clones total ) and eight control mice ( 65 clones total ) were compared , and forebrain DNA from five ethanol-exposed mice ( 33 clones total ) and five control mice ( 43 clones total ) were compared using the Student's t-test . All samples were highly methylated and no differences between the ethanol exposure group and controls were detected ( Figure S2 ) . This suggests that only a subset of IAPs , perhaps those that are usually hypomethylated , are sensitive to ethanol exposure . To detect changes in gene expression genome-wide , we performed expression arrays with liver tissue . The benefit of using liver is its homogeneity; it consists mainly of hepatocytes and consequently subtle changes will be detectable . We compared gene expression between age-matched male mice from the gestational ethanol group ( three samples ) and controls ( four samples ) . In addition to inter-individual variation , some of the genes were consistently differentially expressed in the ethanol-exposed group ( Table S1 ) . Twelve genes were significantly down-regulated ( p<0 . 05 ) in the ethanol-exposed mice . Three of these; LIM domain and actin binding 1 ( Lima1 ) , also known as Eplin , Suppressor of cytokine signaling 2 ( Socs2 ) and CDK5 and Abl enzyme substrate 1 ( Cables1 ) have been associated with growth [29]–[32] . Three; Socs2 , Very low density lipoprotein receptor ( Vldlr ) and Cables1 have been associated with development of the nervous system [33]–[37] and one , Hepcidin antimicrobial peptide ( Hamp1 ) , has been reported to be down-regulated in the livers of alcohol-fed rats [38] . We next focused on identifying the characteristic features of FAS in a/a pups exposed to moderate levels of alcohol in utero . All pups ( from first litters ) were weighed at three weeks of age . It was particularly important not to study Avy mice in these experiments because of the effects on body weight due to Agouti expression . Because litter size is known to influence body weight at weaning , we initially restricted our analysis to litters of 4–5 pups . The gestational ethanol exposure group consisted of 22 offspring , while the control group consisted of 26 offspring . The results ( Figure 3 ) show that the mean weight of offspring of dams that consumed ethanol were significantly lower than that of controls ( Student's t-test , p<0 . 05 ) . A second analysis included litter size as a random effect . Analysis of Variance of weight at 3 weeks , after adjustment for litter size , confirmed that the mean weight of the ethanol exposed group ( n = 73 ) was statistically significantly smaller than the mean weight of the controls ( n = 44 , p<0 . 001 , data not shown ) . The heads of 28–30 day old a/a mice ( seven mice from gestational ethanol exposure group and 10 control mice ) were subjected to micro-computed tomography , and three-dimensional computer-reconstructions at 18 µm resolution were made of each skull . Visual inspection of the reconstructions revealed an obviously smaller skull size in the ethanol group compared to controls . In addition , differences in shape in a few , but not all , individuals in the ethanol group were apparent . Most notable was the marked leftward deviation of the midface in one male ( Figure 4B ) and a significantly reduced interfrontal bone in one female ( Figure 4C ) . To provide more quantitative information on skull shape , the 3D co-ordinates of thirty-four landmarks were recorded for each skull and used in various mathematical-based shape and form analyses . There are two classic approaches in geometric morphometrics: superimposition based methods such as Generalized Procrustes Analysis ( GPA ) [39]–[42] or invariant analyses of shape , such as Euclidean Distance Matrix Analysis ( EDMA ) [43] , [44] . GPA involves translation , rotation and scaling of landmark data through an iterative process during which the distances between the shapes are minimized by applying least-squares criteria . We used GPA to test for the mean shape difference between the groups and to quantify and visualize localized differences in the cranial shape . We also applied the multivariate ordination method Canonical Variates Analyses to the output of the GPA . EDMA , in contrast , uses a coordinate-free ( or invariant ) approach in which all the landmarks are converted into a matrix of inter-landmark distances [44] , [45] . We used EDMA to find the landmark pairs that show the most difference between two groups . Analysis of skull centroid sizes confirmed the observations of statistically significantly reduced cranial size in the ethanol group , even when the smaller body weight is taken into account ( ANOVA , p<0 . 05; Figure 4D ) . Although the severe leftward deviation of the one male skull is biologically highly relevant , we chose to exclude this sample from subsequent shape analyses because of its significant impact on the results . This permitted us to assess the significance of other more subtle changes . However , it was included in the univariate analyses of relative cranial dimensions . In the absence of this outlier , CVA still revealed greater variation in overall craniofacial shape within the ethanol group ( Figure 5A ) . Canonical variate 1 ( CV1 ) clearly separated the females in the ethanol group from other skulls , suggesting a more pronounced effect on female skull shape . Notably , all the females as well as the included males from the ethanol group had positive values for CV2 , whereas the controls spanned both negative and positive values ( see Figure 5A ) , indicative of a similar trend in shape alteration in response to this level of gestational exposure to ethanol . One female from the ethanol group appeared to be unaffected in terms of craniofacial shape and grouped with the control females in all analyses . We then used EDMA to assess the differences in form between the ethanol and control groups . Analysis of the 561 possible inter-landmark measurement combinations assessed by the 34 assigned landmarks ( i . e . 34 ( 34-1 ) /2 ) demonstrate that the majority show a consistent ratio below one , indicative of the fact that they are changed only relative to skull size and do not reflect localized altered shape . Nevertheless , numerous inter-landmark measures were shown to be significantly different from this mean form . The twenty most significant differences ( α = 0 . 1 ) in either direction from the mean form are shown in Figure 5B . Strikingly , almost all of these forty most significant differences pertain to midfacial and palatal inter-landmark measures , highlighting the sensitivity of this region to the ethanol . In particular , these data reveal that the ethanol group as a whole have a relatively wider inter-orbital distance ( inter-landmark measure 7–11 ) , yet relatively shorter midface than controls ( reflected in multiple inter-landmark measures ) . This is consistent with the CVA findings . A univariate analysis of inter-landmark distances ( normalized to centroid size ) also supported these differences between the ethanol and control groups , in particular , confirming the greater relative cranial and inter-orbital width in both males and females compared to the sex-matched controls ( Figure 5C ) . Females from the ethanol group also showed greater variation in ‘nare’ height ( data not shown ) , while males from the ethanol group showed reduced rostrum length ( Figure 5C ) . Although less severe than the changes found with acute ethanol exposure in the mouse [16] , many of these differences are reminiscent of the facial changes seen in individuals presenting the milder end of fetal alcohol spectrum disorder , and support the notion that this ethanol regime provides a useful and relevant model for the effects of ethanol intake in humans .
The Avy allele has been called an epigenetic biosensor for environmental effects on the fetus [46] . Previous studies with this allele have identified a number of nutritional factors or toxic agents that affect expression and epigenetic regulation of Avy in offspring exposed in utero . For example , the addition of methyl supplements or an isoflavone ( called genistein ) to the diet causes hypermethylation of Avy and a shift to a pseudoagouti coat color [3] , [4] , [47] , whereas bisphenol A , a chemical used in the manufacture of polycarbonate plastics causes a shift towards hypomethylation and a yellow coat [48] . We have used the Avy allele to investigate the epigenetic effects of exposure to ethanol , and established two models of moderate exposure in the mouse . The first involves exposure during the first eight days after fertilization; a period that encompasses pre-implantation , implantation and the first two days of gastrulation . This model simulates the effects of ethanol exposure during the first trimester of pregnancy in humans . Based on previous studies [49] we estimated that the peak blood alcohol level in our model is approximately 0 . 12% , which is a realistic human exposure . A World Health Organization ( WHO ) report shows that the maximum legal blood alcohol level for driving in Organization for Economic Cooperation and Development ( OECD ) countries varies from 0 . 02–0 . 08% [50] . The second model involves ethanol consumption for ten weeks immediately prior to fertilization; a period that encompasses multiple cycles of oocyte maturation and ovulation . In mammals , oocyte maturation is characterized by the resumption of meiosis , extrusion of the first polar body and the accumulation of RNAs and proteins in the cytoplasm in preparation for fertilization . Studies of epigenetic reprogramming have shown that following global DNA demethylation , the period of genome-wide remethylation coincides with implantation ( for embryos ) and oocyte growth ( for female germ cells ) [51] . Our results demonstrate that gestational ethanol exposure increases the likelihood of transcriptional silencing at Avy , resulting in an agouti-colored coat . It is worth emphasizing that despite being genetically identical , not all Avy mice become pseudoagouti; rather there is a subtle ∼15% increase in the proportion of pseudoagouti offspring . Previous studies have demonstrated that there is a tight correlation between DNA methylation at Avy and coat color [9] . As expected , bisulfite sequencing showed that the observed coat colors correlated with DNA methylation status in all cases . We did observe atypical hypermethylated clones in five of six yellow mice from the ethanol exposed group that , while not sufficient to change coat color , may reflect a tendency towards increased DNA methylation in this group . Preconceptional ethanol exposure produced a similar shift towards pseudoagouti in Avy offspring . It is likely that the two types of ethanol exposure have different modes of action on Avy because this allele is paternally-derived and not present in unfertilized oocytes . Consequently , the effects of preconceptional ethanol exposure on Avy expression will be indirect , and further work will be required to understand this mechanism . It is of interest that the coat color changes observed in Avy mice exposed to a methyl rich diet can be inherited across generations [52] . It is therefore possible that the altered coat color following alcohol exposure could also be transmitted to the next generation , but was beyond the scope of this study . Our model of moderate gestational ethanol exposure produces a postnatal growth restriction phenotype and craniofacial dysmorphism in line with those seen with FASD in humans . It is possible that the postnatal growth restriction phenotype is an indirect effect; for example , the offspring may be smaller because of deficient maternal care between birth and weaning . It is unlikely that the dam would have been intoxicated or even experiencing ethanol withdrawal symptoms in the postpartum period since ethanol exposure ceased at GD 8 . 5 and water was consumed for the rest of gestation ( ∼11 . 5 days ) and throughout nursing ( 21 days ) . Regardless of whether the phenotype is a direct physiological consequence of exposure in utero or the indirect result of altered maternal behavior , we would argue that it is ultimately a product of exposure to ethanol . Interestingly the effects on skull shape in these mice , like the coat color presentation in Avy mice , are variable despite the fact that the mice are isogenic . Marked variability in phenotype has also been recognized in humans in which not all children of heavily drinking mothers have the typical FAS facial phenotype; the others falling in the continuum of FASDs [14] . While this variability has been attributed to genetic differences , and differences in the level/timing of exposure , it may also be a consequence of stochastic establishment of epigenetic state . The mechanism by which ethanol alters the establishment of epigenetic state at Avy is not known . It has been shown that chronic ethanol consumption can alter DNA methylation by changing the levels of S-adenosylmethionine ( SAM ) , which donates methyl groups to cytosine [53] , [54] . It is also known that chronic or acute ethanol consumption can cause post-translational histone modifications in rat tissues [55]–[59] . The effect of ethanol on the developing embryo has been less studied at the molecular level . Candidate gene and microarray analyses have detected changes in the level of expression ( both up- and down-regulation ) of numerous genes [60]–[62] and decreased global DNA methylation in midgestation embryos has been reported following acute ethanol treatment [24] . Recent studies have also reported altered regulation of several microRNAs by ethanol suggesting a possible role for these RNA species in fetal alcohol syndrome [63] , [64] . Our gestational exposure experiments demonstrate that the epigenome is vulnerable to ethanol during early embryogenesis , a time when the DNA synthetic rate is high and there is genome-wide epigenetic reprogramming . Our preconceptional ethanol exposure experiments show that changes in the maturing oocyte ( another period in development when there are widespread changes to the epigenome ) can also affect offspring phenotype . The identification of microcephaly and midfacial dysmorphism in our gestational exposure model suggests effects on genes other than Avy . Our preliminary genome-wide gene expression analyses of liver in ethanol exposed mice revealed twelve consistently down-regulated and three up-regulated genes . Ongoing work will determine if the expression of these same genes has been changed in other tissues and whether it correlates with alterations in methylation level . The variable and subtle nature of the observed phenotypes will make this work challenging , but our ultimate goal is to gain a better understanding of the molecular processes underlying FASDs .
All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the Animal Ethics Committee of the Queensland Institute of Medical Research ( P986 , A0606-609M ) . The mice used in this study were inbred , genetically identical , C57BL/6J and all environmental factors ( e . g . cage type , environmental enrichment ) were standardized . We chose a voluntary consumption strategy for ethanol exposure instead of intraperitoneal injections or intragastric administration because it produces the least amount of maternal stress . C57BL/6 mice are also known to have a strong drinking preference for 10% ( v/v ) ethanol over water making them ideal for the study [65] , [66] . For gestational ethanol exposure , single mottled Avy/a males were caged with single 6–14 week old a/a females . The majority of females were 6–8 weeks old and virgins in both ethanol-exposured and control groups . The females were checked each morning for a vaginal plug which indicated that mating had taken place . The day of plugging was designated GD 0 . 5 , the male was removed from the cage and the water bottle was replaced with one containing 10% ( v/v ) ethanol . Pregnant females were allowed free access to the drink bottle and food at all times . The ethanol solution was changed and consumption ( ml ) was measured every 24 hours . The average daily consumption was 3 . 1±0 . 4 ml of 10% ( v/v ) ethanol ( or 12 g ethanol/kg body weight/day ) which was not statistically significantly different from the average daily water consumption of control mice ( Student's t-test , two tailed , p = 0 . 8 ) . It has been shown that in female mice , voluntary consumption of 10% ( v/v ) ethanol at 14 g ethanol/kg body weight/day produces an average peak blood alcohol level of ∼120 mg/dl [49] . Only one out of 47 females tested refused to drink ethanol in the initial 24 hour period and was excluded from the analysis . On the final day of exposure , GD 8 . 5 , the ethanol bottle was replaced with a bottle containing water . All dams were subjected to only one cycle of ethanol exposure . Offspring were left with their mothers until weaning ( at 3 weeks of age ) , when their coat color was recorded ( Avy/a mice ) or they were weighed or subjected to micro-computed tomography ( a/a mice ) . For preconceptional ethanol exposure , 6 week old a/a female mice were given 10% ( v/v ) ethanol for 4 days per week ( 4 days ethanol followed by 3 days water ) for ten weeks . After treatment , at 18–22 weeks age , they were mated with mottled Avy/a males . Avy/a offspring were weaned and phenotyped for coat color at three weeks of age . All ethanol and control exposures were performed in parallel so that exactly the same animal house conditions were experienced for all experiments . The coat color of Avy/a offspring was visually classified by a single trained observer ( NK-A ) and placed into one of five categories: yellow ( >95% yellow ) , yellow/mottled ( 75–95% yellow ) , mottled ( 25–74% yellow or 25–74% agouti ) , pseudoagouti/mottled ( 75–95% agouti ) or pseudoagouti ( >95% agouti ) . In the final analysis these categories were combined into three classes: yellow , mottled ( comprised of yellow/mottled , mottled and pseudoagouti/mottled ) and pseudoagouti . For bisulfite sequencing of the Avy allele , 200–400 ng of tail genomic DNA was embedded in agarose and then treated with sodium bisulfite as described previously [10] . The bisulfite-treated DNA was resuspended in 30 µl of water and 5 µl was used in the primary PCR followed by a semi-nested PCR with 2–5 µl of template ( primers were forward 5′ gaaaagagagtaagaagtaagagagagag 3′ , reverse 5′ aaaatttaacacataccttctaaaaccccc 3′ and semi-nested reverse 5′ actccctcttctaaaactacaaaaactc 3′ ) [10] . One bisulfite conversion and PCR was performed for each pseudoagouti sample , while 3–5 independent conversions and 3 PCRs/conversion were performed for each yellow sample . Global IAP LTR sequences were amplified from bisulfite-converted tail and forebrain DNA using universal IAP primers; forward 5′ ttgatagttgtgttttaagtggtaaataaa 3′ and reverse 5′ aaaacaccacaaaccaaaatcttctac 3′ [67] . An agarose-only ( no template ) control was always included and the experiment was only continued if the agarose control was negative at the end of the semi-nested PCR . PCR fragments were gel-isolated and subcloned into the pGEM-T vector ( Promega , Madison , Wisconsin , United States ) . Individually sequenced clones were analyzed with BiQ Analyzer [68] . To avoid bias , clones from the same PCR were only accepted if they differed by either CpG or non-CpG methylation . Any clones with lower than 90% conversion rate were also excluded from the dataset . To detect possible changes in gene expression in gestational ethanol exposure mice compared to the controls , we used the MouseWG-6 v2 . 0 Expression BeadChips ( Illumina ) . We extracted total liver RNA from 28 days old males from control and gestational ethanol groups , using a Qiagen RNeasy Plus-kit ( Qiagen ) . We used a Bioanalyzer ( Agilent RNA 6000 Nano , Agilent ) to confirm the quality of RNA and accepted only samples with RNA Integrity Numbers ( RINs ) above 9 . We amplified RNA using an Illumina TotalPrep RNA Amplification Kit and performed a Whole-Genome Gene Expression Direct Hybridization Assay ( Illumina ) . The gene expression data from scanned microarray images generated by the Illumina BeadArrayTM Reader was analysed by the GenomeStudio Gene Expression Module ( Illumina ) by using probe information . Four control samples from two litters and three gestational ethanol exposure samples from three litters were analysed . Seventeen a/a mice ( ten controls and seven ethanol exposed mice ) aged between 28 and 30 days were subjected to micro-computed tomography using a SkyScan 1076 microtomograph at the Small Animal Tomographic Analysis Facility located at the University of Washington . The sex and treatment breakdown of the microCT samples is female ethanol ( n = 4 ) , female control ( n = 5 ) , male ethanol ( n = 3 ) and male control ( n = 5 ) . Specimens were scanned at 18 micron resolution ( 65 kV , 150 mA , 1 . 0 mm Al filter ) and reconstructed as series of 8-bit grayscale images . Three-dimensional models of the skulls were generated using the thresholding algorithm in Analyze 3D ( Mayo Clinic , version 9 . 0 ) . A grayscale value of 55 was determined to be the optimum threshold value to remove the soft tissue structures and scan noise while keeping the skull morphology intact , and was used for all specimens . Using the point measurement tool of Analyze , 35 landmarks were collected from each specimen ( Text S1 and Figure S3 ) . Specimens were digitized by the same observer ( MM ) to reduce inter-observer error . Visualizations showed that landmark 31 could not be accurately determined in every specimen because of the occasional fusion of the presphenoid and basisphenoid bones . Because geometric morphometrics requires homologous landmarks collected from every specimen , this landmark was omitted in subsequent analyses . Landmark data were fed into various morphometric packages . Using the R statistical package [69] , linear measurements of certain common cranial dimensions were calculated from the landmark coordinates and normalized to their respective skull centroid sizes . Generalized Procrustes Analysis ( GPA ) was also conducted in R by using the SHAPES module . Goodall's F test was used to test for statistical significance of mean shape differences among groups . The Canonical Variates Analysis ( CVA ) was conducted in the MorphoJ package [70] . The loadings of the canonical variates 1 and 2 were used to visualize the cranial shape changes depicted by each axis . The WinEDMA package [71] was used to conduct Euclidean Distance Matrix Analysis . We used the FORM procedure of WinEDMA to find the landmark pairs that significantly differed between two mean forms ( i . e . , ethanols and controls ) as measured by the form difference matrix . Following Lele and Richtsmeier [45] , the 90% confidence intervals for the pairwise ratios were calculated by bootstrapping the form difference matrix 1000 times . | In humans it has been known for some time that exposure to environmental insults during pregnancy can harm a developing fetus and have life-long effects on the individual's health . A well known example is fetal alcohol syndrome , where the children of mothers that consume large amounts of alcohol during pregnancy exhibit growth retardation , changes to the shape and size of the skull , and central nervous system defects . At present the molecular events underlying fetal alcohol syndrome are unknown . We have developed a model of alcohol exposure in the mouse , in which the genetics and the environment can be strictly controlled . We find that chronic exposure of the fetus to a physiologically relevant amount of alcohol during the first half of pregnancy results in epigenetic changes at a sensitive reporter gene and produces fetal alcohol syndrome-like features in some mice . Our model is a useful tool to study the underlying causes of fetal alcohol syndrome , and our work raises the interesting possibility that the long-term physical effects of alcohol exposure during pregnancy are mediated by epigenetic changes established in the fetus and then faithfully remembered for a lifetime . In the future , such epigenetic changes could be used as markers for the preclinical diagnosis and treatment of fetal alcohol spectrum disorders . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/disease",
"models",
"molecular",
"biology/dna",
"methylation",
"genetics",
"and",
"genomics/epigenetics",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] | 2010 | Maternal Ethanol Consumption Alters the Epigenotype and the Phenotype of Offspring in a Mouse Model |
We recently showed that the exocytosis regulator Synaptotagmin ( Syt ) V is recruited to the nascent phagosome and remains associated throughout the maturation process . In this study , we investigated the possibility that Syt V plays a role in regulating interactions between the phagosome and the endocytic organelles . Silencing of Syt V by RNA interference revealed that Syt V contributes to phagolysosome biogenesis by regulating the acquisition of cathepsin D and the vesicular proton-ATPase . In contrast , recruitment of cathepsin B , the early endosomal marker EEA1 and the lysosomal marker LAMP1 to phagosomes was normal in the absence of Syt V . As Leishmania donovani promastigotes inhibit phagosome maturation , we investigated their potential impact on the phagosomal association of Syt V . This inhibition of phagolysosome biogenesis is mediated by the virulence glycolipid lipophosphoglycan , a polymer of the repeating Galβ1 , 4Manα1-PO4 units attached to the promastigote surface via an unusual glycosylphosphatidylinositol anchor . Our results showed that insertion of lipophosphoglycan into ganglioside GM1-containing microdomains excluded or caused dissociation of Syt V from phagosome membranes . As a consequence , L . donovani promatigotes established infection in a phagosome from which the vesicular proton-ATPase was excluded and which failed to acidify . Collectively , these results reveal a novel function for Syt V in phagolysosome biogenesis and provide novel insight into the mechanism of vesicular proton-ATPase recruitment to maturing phagosomes . We also provide novel findings into the mechanism of Leishmania pathogenesis , whereby targeting of Syt V is part of the strategy used by L . donovani promastigotes to prevent phagosome acidification .
Phagocytosis consists in the uptake and destruction of invading microorganisms , thereby playing an essential role in host defense against infection [1] . Following internalization , microbes end up in a vacuole , the phagosome , which engages in a maturation process involving highly regulated fusion and fission events with early and late endosomes , and with lysosomes [2] , [3] . This leads to the acidification of the phagosome and the acquisition of an array of hydrolases , culminating in the generation of a highly microbicidal environment [4] . Soluble N-ethylmaleimide-sensitive factor protein attachment protein receptor ( SNARE ) -mediated membrane fusion events regulate phagosome maturation by facilitating interactions with the endocytic compartments [5] . Hence , VAMP3 and syntaxin 13 are present transiently on the young phagosome to regulate early maturation steps , whereas VAMP7 and syntaxin 7 remain associated with the phagosome to regulate interactions with late endosomes/lysosomes [6]–[8] . The lysosome-associated Synaptotagmin ( Syt ) VII , which controls membrane delivery to nascent phagosomes [9] , is also involved in phagolysosome fusion [9] , [10] . Other components and partners of these SNARE fusion machineries required during phagosome maturation remain to be identified . Phagolysosome biogenesis is an important means of controling microbial growth . Yet , several pathogenic microorganisms have evolved mechanisms to subvert the phagosome maturation process , thus avoiding an encounter with the macrophage microbicidal machinery including exposition to reactive oxygen species and to acidification [4] , [11] , [12] . Protozoan parasites of the genus Leishmania cause a spectrum of diseases in humans , ranging from self-healing ulcers to potentially fatal visceral leishmaniasis , which affect millions of people worldwide . Leishmania is transmitted to mammals under its promastigote form during the bloodmeal of infected sand flies . Following phagocytosis by macrophages , promastigotes must avoid destruction to differentiate into amastigotes , the mammalian stage of the parasite that replicate inside acidic and hydrolase-rich parasitophorous vacuoles [13]–[15] . To avoid the microbicidal arsenal of macrophages , L . donovani and L . major promastigotes create an intracellular niche through the inhibition of phagolysosome biogenesis [16]–[19] . Genetic and biochemical approaches established that this inhibition is strictly dependent on the presence of lipophosphoglycan ( LPG ) , an abundant surface glycolipid consisting of a polymer of Galβ1 , 4Manα1-PO4 units anchored into the promastigote membrane via an unusual glycosyl phosphatidylinositol [20] , [21] . Hence , phagosomes harboring LPG-defective promastigotes quickly mature into functional phagolysosomes and coating of the Galβ1 , 4Manα1-PO4-defective mutant lpg2-KO with purified LPG conferred the capacity to inhibit phagosome-lysosome fusion [16] , [17] , [19] , [22] , [23] . LPG-mediated phagosome remodeling is characterized by a periphagosomal accumulation of F-actin [22] , [23] and by the exclusion of cytosolic components of the NADPH oxidase from the phagosome membrane [24] . By creating an environment devoid of oxidants , L . donovani promastigotes evade a major microbicidal mechanism of macrophages and can initiate their differentiation into amastigotes . The ability of LPG to inhibit phagosome maturation is consistent with its role in the establishment of L . donovani and L . major promastigotes inside macrophages [24] , [25] . A possible mechanism by which LPG exerts its action on phagosome maturation involves the transfer of LPG from the parasite surface to lipid microdomains present in the phagosome membrane , causing a disorganization of these structures and preventing their formation after phagocytosis [26]–[29] . Phagosomal lipid microdomains are essential for the recruitment/assembly of the NADPH oxidase and the vacuolar proton-ATPase and are involved in the regulation of phagosome-endosome fusions [27] , [30] , [31] . Disruption of lipid microdomains by microbial virulence factors is likely to facilitate the establishment of infection through an effect on phagolysosomal biogenesis , as described for the cyclic β-1 , 2-glucans of Brucella abortus and the lipoarabinomannan of Mycobacterium tuberculosis [32] , [33] . How lipid microdomains regulate interactions between phagosomes and the endocytic system is unclear . The fact that proteins involved in membrane fusion such as SNAREs and Syts are located in lipid microdomains is consistent with these structures acting as fusion sites [34] , [35] . Recently , we identified the exocytosis regulator Syt V [36]–[39] as a recycling endosome-associated protein that is recruited to the forming phagosome independently of the phagocytic receptor engaged [40] . Silencing of Syt V by RNAi revealed a role for this protein during phagocytosis , particularly under conditions of high membrane demand , possibly through the mobilization of recycling endosomes as a source of endomembrane . The association of Syt V with the phagosome throughout the maturation process raised the possibility that Syt V regulates interactions with the endocytic system [40] . Here , we provide evidence for a novel function of Syt V in phagolysosome biogenesis , where it controls the acquisition of cathepsin D and the vesicular proton-ATPase . We also provide novel insight into the mechanism of L . donovani pathogenesis with the demonstration that insertion of LPG into GM1-containing microdomains impairs the association of Syt V to phagosome membranes , enabling L . donovani promatigotes to inhibit the recruitment of the vesicular proton-ATPase to phagosomes , thereby preventing their acidification .
Syt V , a regulator of exocytosis , is recruited to the nascent phagosome and remains associated throughout the maturation process [40] , suggesting that it may participate in the regulation of phagolysosome biogenesis . Maturing phagosomes sequentially interact with various endocytic organelles to acquire hydrolases such as cathepsins and the proton-vacuolar ATPase ( V-ATPase ) , which is responsible for phagosome acidification [2] , [41] , [42] . To assess the potential role of Syt V in the acquisition of microbicidal features , we inhibited its expression by transfecting RAW 264 . 7 cells with a siRNA to Syt V [40] ( Figure 1A ) and we examined the localization of phagosomal markers following the internalization of Zymosan ( Zym ) or latex beads . Our results show that in the absence of Syt V , recruitment of both the early endosomal ( EEA1 ) and the lysosomal ( LAMP1 ) markers to Zym-containing phagosomes was normal ( Figures 1B and S1A and B ) , whereas the acquisition of cathepsin D and the V-ATPase c subunit was inhibited ( Figure 1B–E ) . Reduction in cathepsin D acquisition ranged from 25 to 35% for phagosomes containing beads and from 41 to 48% for phagosomes containing Zym , in five independent experiments . In the case of the V-ATPase c subunit , the reduction ranged from 30 to 50% for phagosomes containing beads and from 45 to 60% for phagosomes containing Zym in five independent experiments ( Figure 1C–E ) . Interestingly , silencing of Syt V had no detectable effect on the acquisition of cathepsin B ( Figure 1C ) . These results provide evidence that Syt V selectively regulates the phagosomal acquisition of cathepsin D and the V-ATPase c subunit . Given their ability to inhibit phagosome maturation [16] , [17] , [24] , we explored the impact of L . donovani promastigotes and their LPG on the phagosomal association of Syt V . Accordingly , we infected the mouse macrophage cell line RAW 264 . 7 stably expressing a Syt V-GFP fusion protein ( Syt V-GFP RAW 264 . 7 cells ) with either wild-type ( WT ) L . donovani promastigotes , the LPG-defective lpg1-KO mutant , the Galβ1 , 4Manα1-PO4-defective lpg2-KO mutant or the lpg2-KO add-back ( lpg2-KO+LPG2 ) . We used Zym as a positive control for the recruitment of Syt V to phagosomes [40] . Our results show that Syt V-GFP was present on over 80% of phagosomes containing either lpg1-KO promastigotes , lpg2-KO promastigotes , or Zym ( Figure 2A and B ) . In contrast , we detected Syt V-GFP on 54 to 65% of phagosomes containing either WT or lpg2-KO+LPG2 promastigotes in three independent experiments . Quantification analyses revealed a three-fold reduction in the levels of Syt V-GFP present on those positive phagosomes with respect to phagosomes containing either lpg1-KO or lpg2-KO promastigotes ( Figure 2C ) . These observations suggested that LPG impairs the phagosomal recruitment of Syt V . To directly address the impact of LPG on the recruitment of Syt V to phagosomes , we fed bone marrow-derived macrophages ( BMM ) with either Zym or Zym coated with purified LPG ( LPG-Zym ) [22] . Consistent with previous observations [17] , [22] , we found a reduced acquisition of LAMP-1 on phagosomes containing LPG-Zym , whereas the recruitment of EEA1 to phagosomes containing Zym or LPG-Zym was similar ( Figure 3A and B ) . In the case of Syt V , we detected its presence on 24 to 30% of phagosomes containing LPG-Zym compared to over 60% of phagosomes containing Zym at all time points tested in three independent experiments ( Figure 3C and D ) . Quantification analyses showed that the levels of Syt V present on those positive phagosomes containing LPG-Zym was significantly lower than the Syt V levels on phagosomes containing Zym ( Figure 3C and D ) . We obtained similar results with the Syt V-GFP RAW 264 . 7 cells ( Figures 3E and S2 ) . Furthermore , the signals for Syt V ( green ) and LPG ( red ) rarely superimposed on the phagosome membrane ( Figure 4A ) , and fluorescence intensity line scans acquired along the periphery of phagosomes showed that the most intense LPG and Syt V signals never overlapped , at both 30 min and 120 min after the initiation of phagocytosis ( Figure 4B ) . We made similar observations in Syt V-GFP RAW 264 . 7 cells ( Figure 4C and D ) . Collectively , these results established that insertion of LPG into the phagosomal membrane caused the exclusion of Syt V in a very localized manner . In rat brain synaptosomes , a fraction of Syt I and Syt II is present in lipid rafts [34] . To examine whether LPG-mediated exclusion of Syt V from phagosomes was related to the insertion of LPG into lipid microdomains [27] , [29] ( Figure 5D ) , we first determined whether phagosome-associated Syt V was present in these microdomains . Our results clearly show that a fraction of Syt V colocalizes with GM1-microdomains on Zym-containing phagosomes ( Figure 5A , arrowheads ) . Consistently , cholesterol depletion by methyl-β-cyclodextrin inhibited the recruitment of Syt V ( Figure 5B and C ) . Having established that phagosomal Syt V associates with GM1-containing microdomains , we examined the localization of LPG , Syt V , and GM1 on phagosomes containing either Zym or LPG-Zym . For phagosomes containing Zym , the signals for Syt V ( blue ) and GM1 ( red ) superposed to a large extent and fluorescence intensity line scans acquired along the periphery of a representative phagosome showed that most of the Syt V and GM1 signals overlapped ( Figure 5E and F , top panel ) . In contrast , on phagosomes containing LPG-Zym , the signals for LPG and GM1 colocalized , whereas most of the remaining Syt V signal was not associated with GM1 ( representative phagosome , Figure 5E and F , bottom panel ) . These results established that association of LPG with GM1-containing microdomains resulted in the exclusion or dissociation of Syt V from the phagosome membrane . The demonstration that Syt V regulates acquisition of the V-ATPase led us to verify the hypothesis that exclusion or dissociation of Syt V from phagosomes containing L . donovani promastigotes may impair the recruitment of the V-ATPase to these phagosomes . At 2 h after the initiation of phagocytosis , our results from three independent experiments showed a reduction in the recruitment of the V-ATPase c subunit on phagosomes containing WT promastigotes , ranging from 54 to 62% with respect to phagosomes containing either lpg1-KO or lpg2-KO promastigotes ( Figure 6A and B ) . Co-localization of the V-ATPase c subunit with LAMP-1 on phagosomes containing lpg1-KO promastigotes showed that the V-ATPase c subunit was present on the phagosome membrane ( Figure S3 ) . As expected , phagosomes containing lpg2-KO+LPG2 cells were similar to WT-phagosomes with respect to the presence of the V-ATPase . We next monitored the acidification of L . donovani promastigote-containing phagosomes using the lysosomotropic agent LysoTracker red as an indicator of phagosome pH . Our results showed a clear correlation between the presence of the V-ATPase c subunit and the association of LysoTracker red to phagosomes ( Figure 6C ) . In Figure 1 , we showed that silencing of Syt V inhibited recruitment of the V-ATPase c subunit to phagosomes containing Zym or latex beads . In Figure 6D , we show that silencing of Syt V abrogated recruitment of the V-ATPase c subunit to phagosomes containing lpg1-KO and lpg2-KO mutants . In the case of phagosomes containing either WT or lpg2-KO+LPG2 promastigotes , Syt V silencing had the same effect as the presence of LPG on the recruitment of the V-ATPase c subunit ( Figure 6D ) . Collectively , these results show that LPG enables L . donovani promatigotes to inhibit phagosomal recruitment of the V-ATPase by a Syt V-dependent mechanism and to prevent acidification . Remarkably , at 24 h after the initiation of phagocytosis , we detected the V-ATPase c subunit on only 10 to 17% of phagosomes containing L . donovani promastigotes in three independent experiments , consistent with LPG still being present ( Figure 7A and C ) . At this time point , we detected LysoTracker red on only 20% of phagosomes containing WT promastigotes ( not shown ) , indicating that promastigotes remodel their intracellular niche to establish infection in a compartment that fails to acidify , at a time when differentiation into amastigotes takes place . In contrast , we detected the V-ATPase c subunit on 66 to 71% of phagosomes containing L . donovani amastigotes at both 2 h and 24 h after the initiation of phagocytosis ( Figure 7B and C ) . This observation is consistent with the fact that amastigotes replicate in an acidic phagolysosomal compartment [14] .
The exocytosis regulator Syt V is recruited to the nascent phagosome and remains associated throughout the maturation process [40] , leading us to investigate its potential role in modulating interactions between the phagosome and endocytic organelles . Our results revealed that whereas silencing of Syt V had no effect on the recruitment of EEA1 , LAMP-1 , and cathepsin B , it inhibited the phagosomal acquisition of cathepsin D and of the V-ATPase c subunit . These findings indicated that Syt V plays a role in phagolysosome biogenesis , possibly by regulating the interaction between phagosomes and a subset of late endosomes or lysosomes enriched in cathepsin D and in the V-ATPase c subunit . Alternatively , Syt V may be needed to reach the level of phagosome maturation necessary to acquire the machinery that regulates the recruitment of cathepsin D and the V-ATPase c subunit . Our finding that acquisition of cathepsin B and cathepsin D is mediated by distinct mechanisms supports the demonstration that various hydrolases appear sequentially , at various time points during phagosome maturation [42] . This view is also consistent with evidence that various sub-populations of early endosomes , late endosomes , and lysosomes co-exist and that these compartments contain significant heterogeneity [43] . Together with previous findings [27] , our results show that phagosomal acquisition of the V-ATPase and LAMP-1 are mediated through distinct mechanisms . Hence , the observations that LAMP-1 is recruited to phagosomes independently of Syt V and that L . donovani promastigotes ( and LPG ) impair the recruitment of LAMP-1 point to the existence of other inhibitory mechanisms and illustrate the complexity of phagolysosome biogenesis . The role of Syt V in regulating interactions between the phagosome and the endosomal compartments thus seems specific and further studies will be necessary to understand its precise role during phagosome maturation . Recent studies by Andrews and colleagues revealed that the lysosome-associated Syt VII , which controls membrane delivery to nascent phagosomes [9] , is involved in phagolysosome fusion [9] , [10] . It will be of interest to determine whether Syt V and Syt VII use similar mechanisms to regulate phagolysosome biogenesis . To establish infection inside macrophages , L . donovani promastigotes , the form of the parasite transmitted to mammals by the sand fly vector , create an intracellular niche by inhibiting phagolysosome biogenesis [16] . Genetic and biochemical approaches revealed that this inhibition is mediated by the parasite surface glycolipid LPG [16] , [17] , [22] . Insight into the mechanism of this inhibition came from the observations that LPG transfers from the parasite surface to the nascent phagosome membrane [26] , where it disrupts existing lipid microdomains and alters the formation of these structures after promastigote internalization [28] , [29] . Whereas the precise mechanism remains to be elucidated , the current model is that LPG inserts into lipid microdomains via its GPI anchor , thereby allowing the negatively charged Galβ1 , 4Man-PO4 polymer of LPG to directly interfere with the clusterization of molecules into these microdomains . This model is consistent with the demonstration that alteration of membrane properties is dependent on the length of the Galβ1 , 4Man-PO4 polymer [16] , [44] . Because of their role in clustering specific sets of proteins , membrane lipid microdomains are central to a wide variety of cellular processes , including regulated exocytosis [45] , [46] . Our findings that Syt V was present in GM1-enriched phagosome microdomains and that LPG inserts into or associates with these structures to interfere with the phagosomal association of Syt V thus provides new insight into the mechanism of LPG-mediated inhibition of phagolysosome biogenesis . Acquisition of an array of hydrolases and acidification of the phagosome enable the generation of a highly microbicidal environment [4] and the creation of a compartment competent for antigen processing and presentation [47] . To circumvent killing following uptake by macrophages , several intracellular microorganisms interfere with phagosome acidification and maturation [4] , [12] , [48] . The discovery that L . donovani promastigotes establish infection inside a compartment from which the V-ATPase is excluded may thus be favorable for parasite survival . Incidentally , a recent study showed that phagosome acidification is defective in Stat1−/− macrophages and this correlated with an increased survival of L . major promastigotes , suggesting a role for acidic pH in the control of intracellular Leishmania growth early during infection [49] . Furthermore , the finding that phagosomes containing L . donovani promastigotes fail to acquire the V-ATPase and acidify even at 24 hours post-infection provides new insight on our undestanding of Leishmania biology . Indeed , in the absence of data on the pH of promastigote-containing phagosomes , it has been assumed that promastigotes initiate infection in an acidic environment and that differentiation of promastigotes into amastigotes is mainly triggered by a rapid exposure to an acidic environment and elevated temperature [50] . Exclusion of the V-ATPase raises the possibility that L . donovani promastigotes initiate the differentiation process in a non-acidified environment . Further studies will be required to fully address this point . An issue that remains unsolved pertains to the acquisition of phagolysosomal features and acidification of parasite-containing vacuoles upon completion of the differentiation of promastigotes into amastigotes . Indeed , previous work by Antoine and colleagues [14] established that L . amazonensis amastigotes reside within an acidic vacuole ( pH 4 . 7–5 . 2 ) , in agreement with the notion that Leishmania amastigotes are internalized within a vacuole that rapidly acquires lysosomal features and in which amastigotes proliferate [13] , [51] . Consistent with these previous reports , we showed the presence of LAMP-1 and the V-ATPase c subunit on phagosomes containing L . donovani amastigotes as early as 2 h after internalization . A possible explanation is that during the first few days post-infection , the presence of LPG in the phagosome membrane prevents acidification and maturation , allowing promastigote-to-amastigote differentiation to take place . The down-regulation of LPG biosynthesis below detectable levels in amastigotes [52] may enable phagosomes to gradually acquire lysosomal features and to acidify . Little is known on the mechanisms that regulate recruitment of the V-ATPase to maturing phagosomes . The identification of Syt V as a regulator of this process and the fact that Syt V is present in microdomains of the phagosome membrane is consistent with the notion that these structures are important for the recruitment of the V-ATPase to the phagosome membrane [27] . Of interest , the V-ATPase c subunit has been previously identified in Triton X-100-resistant fractions from rat brain synaptic vesicles in association with synaptobrevin 2 and synaptophysin [53] , leading the authors of that study to conclude that this interaction may play a role in recruiting the V-ATPase to synaptic vesicles . Whether Syt V is part of such a SNARE complex on phagosomes and the characterization of this complex are important issues that await further investigation . In this study , we provided novel findings into the mechanism of Leishmania pathogenesis , whereby targeting of Syt V , which plays a role in the acquisition of phagosome microbicidal properties , is part of the strategy used by L . donovani promastigotes to create a niche propitious to the establishment of infection within mammalian hosts ( see working model , Figure 8 ) . Interestingly , phagocytosis of either zymosan or lpg2-KO promastigotes coated with the virulence glycolipid lipoarabinomannan from Mycobacterium tuberculosis , impaired the phagosomal association of Syt V ( Figure S4 ) . Whether other intracellular microorganisms use a similar mechanism to remodel their intracellular niche remains to be investigated .
All animals were handled in strict accordance with good animal practice as defined by the Canadian Council on Animal Care , and all animal work was approved by the Comité institutionel de protection des animaux of INRS- Institut Armand-Frappier ( protocol 0811-08 ) . BMM were obtained by growing bone marrow cells from female BALB/c mice at 37°C in 5% CO2 for 7 days in Dulbecco Modified Eagle Medium with L-glutamine ( Life Technologies ) supplemented with 10% heat-inactivated FBS ( Hyclone , Logan , UT ) , 10 mM Hepes ( pH 7 . 4 ) and antibiotics ( complete medium ) in the presence of 15% ( v/v ) L929 cell-conditioned medium as a source of colony-stimulating factor ( CSF ) -1 [54] . BMM were made quiescent by culturing them in the absence of CSF-1 for 18 h prior to being used . The murine macrophage cell line RAW 264 . 7 was grown in complete medium in a 37°C incubator with 5% CO2 . Stably transfected RAW264 . 7 cells expressing Syt V-GFP ( Syt V-GFP RAW 264 . 7 cells ) were previously described [40] . Transfectants were cultured in complete medium containing 500 µg/ml G418 ( Life Technologies ) . Leishmania donovani promastigotes ( Sudanese strain 1S ) were grown at 26°C in RPMI 1640 medium supplemented with 20% heat-inactivated FBS , 100 µM adenine , 20 mM 2-[N-morpholino]ethanesulphonic acid ( pH 5 . 5 ) , 5 µM hemin , 3 µM biopterin , 1 µM biotin and antibiotics . The isogenic L . donovani LPG-defective mutants lpg1-KO and lpg2-KO were described previously [55] . The lpg1-KO mutant secretes repeating Galβ1 , 4Manα1-PO4-containing molecules , but lacks the ability to assemble a functional LPG glycan core [56] , precluding synthesis of LPG . The lpg2-KO mutant expresses the truncated LPG Gal ( α1 , 6 ) Gal> ( α1 , 3 ) Galf ( β1 , 3 ) [Glc ( α1-P ) ]Man ( α1 , 3 ) Man ( α1 , 4 ) GN ( α1 , 6 ) -PI , and does not synthesize repeating Galβ1 , 4Manα1-PO4 units [57] . The lpg2-KO+LPG2 add-back was grown in the presence of 50 µg/ml G418 . For infections , promastigotes were used in late stationary phase of growth . L . donovani amastigotes ( Strain LV9 ) were isolated from the spleen of infected female LVG Golden Syrian hamsters ( Charles River , St-Constant , QC , Canada ) , as described [58] . The rabbit anti-Syt V spacer antiserum was raised against the cytoplasmic region between the transmembrane and the C2 domain ( aa 71–216 ) [37] and was affinity-purified . The rat monoclonal antibody against LAMP-1 developed by J . T . August ( 1D4B ) was obtained through the Developmental Studies Hybridoma Bank at the University of Iowa , and the National Institute of Child Health and Human Development . The rabbit antiserum against the 16 kDa proteolipid subunit ( c subunit ) of the V0 sector of the V-ATPase was kindly provided by Dr . Mhairi Skinner ( University of Guelph , ON , Canada ) [59] . The mouse monoclonal antibody against EEA1 was from BD Transduction Laboratories . The rabbit antiserum against cathepsin B was from Millipore and the rabbit antiserum against cathepsin D was from Upstate . The mouse monoclonal anti-LPG ( CA7AE ) was prepared from hybridoma supernatant [60] . Methyl-β-cyclodextrin ( MβCD ) was from Sigma ( St-Louis , MO , USA ) . LPG was isolated from the log phase cultures of L . donovani promastigotes as previously described [61] , [62] . Purified lipoarabinomannan ( LAM ) from H37Rv strain of Mycobacterium tuberculosis was from Colorado State University ( Fort Collins , CO , USA ) . Syt V silencing by RNAi was performed as previously described [40] using a small interfering RNA ( siRNA ) corresponding to nucleotides 94–112 of the Syt V cDNA [38] , whereas a siRNA specific to GFP was used as a negative control [63] . Adherent RAW 264 . 7 cells were transfected with siRNA duplexes at a final concentration of 240 nM using OligoFectamine ( Invitrogen ) as described [63] . A BLAST search against the mouse genome sequence database was performed to ensure that the chosen siRNA sequences targeted only the mRNA of interest . Cholesterol depletion was achieved by incubating macrophages with 10 mmol/L methyl-β-cyclodextrin ( MβCD ) ( Sigma ) in serum-free medium at 37°C for 1 h . Cells were washed with PBS before particle internalization . Purified LPG and LAM were sonicated and added to the particles at a final concentration of 25 µM in PBS , pH 7 . 3 , incubated at 37°C for 1 h . Particles were washed and resuspended in complete medium prior to phagocytosis experiments . The efficiency of LPG coating was assessed by immunofluorescence using the anti-repeating unit antibody CA7AE . Complement opsonization of L . donovani promastigotes was done as described [23] and complement opsonisation of beads and zymosan was carried out by incubating the particles in DMEM supplemented with 10% mouse serum for 30 min at 37°C prior to phagocytosis . For synchronized phagocytosis assays , macrophages were incubated with particles at a particle-to-cell ratio of 15∶1 ( unless otherwise specified ) for 15 min at 4°C . Excess particles were removed by several thorough washes with DMEM and phagocytosis was triggered by transferring the cells to 37°C for the indicated time points before processing for microscopy . Macrophages were fixed for 10 min in PBS containing 2% paraformaldehyde , permeabilized using 0 . 1% Triton X-100 , and nonspecific binding to surface FcγR was blocked using 1% BSA , 2% goat serum , 6% milk , and 50% FBS . For immunostaining , cells were labeled with the appropriate combinations of primary antibodies or antisera ( anti-Syt V , LAMP-1 , EEA1 , cathepsin D , cathepsin B , V-ATPase , LPG ) , and secondary antibodies ( anti-rabbit , anti-mouse or anti-rat AlexaFluor 488 , 568 or 647; Molecular Probes ) . DRAQ5 ( Biostatus , Leicestershire , UK ) was used to visualize macrophage and parasite nuclei and CTX-B-568 or 647 ( Molecular Probes ) was used to visualize GM1-enriched rafts . Syt V-GFP RAW 264 . 7 cells were fixed and directly incubated with DRAQ5 before being mounted or subjected to immunofluorescence . Of note , we used Syt V-GFP RAW 264 . 7 cells to localize Syt V following infection with L . donovani promastigotes because our antiserum against Syt V cross-reacts with Leishmania epitopes . All coverslips were mounted on glass slides with Fluoromount-G ( Southern Biotechnology Associates ) . Detailed analysis of protein presence and localization on the phagosome was performed using an oil immersion Nikon Plan Apo 100 ( N . A . 1 . 4 ) objective mounted on a Nikon Eclipse E800 microscope equipped with a Bio-Rad Radiance 2000 confocal imaging system ( Bio-Rad , Zeiss ) . Images were obtained using appropriate filters , through the sequential scanning mode of the LaserSharp software ( Bio-Rad Laboratories , Zeiss ) with a Kalman filter of at least 6 . BMM were preloaded with the acidotropic dye LysoTracker Red ( Molecular Probes , Eugene , OR ) diluted in DMEM ( 1∶1000 ) for 2 h at 37°C . Cells were washed and infected with promastigotes for 2 h at 37°C as described in Phagocytosis assay . Cells were then rinsed , fixed with 2% paraformaldehyde for 10 min , washed and directly incubated 20 min with DRAQ5 before being mounted for confocal analysis . To assess the recruitment of proteins of interest , we assessed the presence or absence of staining on the phagosome membrane for each protein , and at least 100 phagosomes were randomly scanned for each condition . To quantify the levels of Syt V and Syt V-GFP ( Figures 2C , 3D and 3E ) , EEA1 ( Figure 3A ) or LAMP-1 ( Figure 3B ) , we determined the relative staining intensity as follows . The 488 and 568 nm excitation channels ( emission 515/30 and 600/40 respectively ) were separated and the protein staining rim around each phagosome was manually traced with a one pixel width . The fluorescence intensity of individual pixels was determined using the software Image J and an average intensity was calculated for each fluorescence intensity profile . To normalize intensity values of all phagosomes , cytosol intensity was also evaluated in the proximity area of the phagosome under study but far enough from the phagosome membrane to avoid quantifying residual phagosome fluorescence . Final phagosome intensity was expressed as the ratio of phagosome intensity ( P ) on cytosol intensity ( C ) , thus P/C . In all cases , we ensured that signal intensity was not at saturation and the 20 more intense staining for each condition were selected and the average compared for the intensity level of each protein . Statistical analyses were performed using Student's two-tail two-sample unequal variance test . | Upon their internalization by macrophages , Leishmania donovani promastigotes inhibit phagolysosome biogenesis . This inhibition is mediated by the virulence glycolipid lipophosphoglycan ( LPG ) , attached to the promastigote surface . We recently showed that the exocytosis regulator Synaptotagmin ( Syt ) V controls early steps of phagocytosis , and remains associated to the phagosome during the maturation process . Here , we show that Syt V contributes to phagolysosome biogenesis by regulating the acquisition of the hydrolase cathepsin D and the vesicular proton-ATPase . Insertion of LPG into lipid microdomains of the phagosome membrane excluded Syt V from phagosomes , enabling L . donovani promatigotes to inhibit the recruitment of the vesicular proton-ATPase to phagosomes , preventing their acidification . Collectively , our results provide novel insight into the mechanism of vesicular proton-ATPase recruitment to maturing phagosomes and reveal how the virulence glycolipid LPG contributes to the mechanism of L . donovani pathogenesis by preventing phagosome acidification . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunology/leukocyte",
"development",
"immunology/innate",
"immunity",
"cell",
"biology/membranes",
"and",
"sorting",
"infectious",
"diseases/protozoal",
"infections",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
] | 2009 | The Leishmania donovani Lipophosphoglycan Excludes the Vesicular Proton-ATPase from Phagosomes by Impairing the Recruitment of Synaptotagmin V |
Poxviruses include medically important human pathogens , yet little is known about the specific cellular factors essential for their replication . To identify genes essential for poxvirus infection , we used high-throughput RNA interference to screen the Drosophila kinome for factors required for vaccinia infection . We identified seven genes including the three subunits of AMPK as promoting vaccinia infection . AMPK not only facilitated infection in insect cells , but also in mammalian cells . Moreover , we found that AMPK is required for macropinocytosis , a major endocytic entry pathway for vaccinia . Furthermore , we show that AMPK contributes to other virus-independent actin-dependent processes including lamellipodia formation and wound healing , independent of the known AMPK activators LKB1 and CaMKK . Therefore , AMPK plays a highly conserved role in poxvirus infection and actin dynamics independent of its role as an energy regulator .
In order to successfully infect cells , viruses must remodel the cellular environment to allow for the reallocation of resources to viral production . Poxviruses are large double stranded ( ds ) DNA viruses that have a sophisticated lifecycle characterized by a number of temporally regulated steps . Vaccinia virus is the prototypical poxvirus , was used as the vaccine to eradicate smallpox , and has been the most thoroughly characterized [1] . To initiate infection , vaccinia first binds , enters cells , uncoats , and expresses early gene products . Next , genomic DNA replication occurs , followed by intermediate and late gene expression . Assembly , maturation , and virus release completes the cycle . Although poxviruses encode a large number of genes ( >200 ) , they remain obligate intracellular pathogens and require a multitude of activities hijacked from their host cell . While many viral factors required for various steps in the vaccinia lifecycle have been described , the specific host factor contribution is less clear . In particular , an early step in the infection cycle involves cell penetration . This step is critical for the initial establishment of infection , and also presents a good target for anti-viral therapeutics [2] . Different families of viruses have developed diverse strategies for entering cells; some fuse at the plasma membrane , while others co-opt one of the many cellular endocytic routes [3] . Studies have demonstrated that macropinocytosis is an important endocytic route of vaccinia entry [4] . Generally , macropinocytosis is a nonselective route for bulk fluid-phase uptake and is not constitutively active , but is induced by growth factors , and also by some pathogens including vaccinia [5] [6] . This active endocytic process induces extensive actin cytoskeletal rearrangement , leading to membrane ruffling , lamellipodia formation , and the internalization of extracellular fluid and membrane . Consistent with this , vaccinia entry is dependent upon modulation of the actin cytoskeleton , and initiates macropinocytosis by inducing dramatic actin-rich microvilli protrusions followed by global myosin II-dependent blebbing , thereby promoting virus uptake [4] [7] . Induction triggers the activation of receptor tyrosine kinases ( RTKs ) which activate complex signaling cascades leading to the induction of these actin extensions which extend the plasma membrane allowing fluid-phase capture . This process involves signaling cascades that converge on members of the Ras superfamily of GTPases in particular , Rab5 and Rac1 [6] [8] . Rac1 contributes to a number of cellular processes that require extensive actin dynamics , and its signaling is carefully regulated by several guanine exchange factors as well as by crosstalk with other Rho family GTPases [9] [10] [11] . Again , as for growth factor dependent macropinocytosis , vaccinia-induced uptake is dependent upon Rac1 [4] [7] . Additional kinases such as p21-activated kinase ( PAK1 ) are then activated along with actin-associated proteins that lead to large-scale actin rearrangements , lipid modifications , and ultimately macropinosome formation [4] [5] . While some specific kinase families have been implicated in macropinocytosis ( e . g . , protein kinase C ( PKC ) , serine/threonine kinases , tyrosine kinases , and phosphatidylinositol kinases ) [5] , many of the specific factors have not been identified , and in some cases the specific role of factors such as PKC , is not well understood . Therefore , there are many additional cellular signaling factors remaining to be identified for this complicated uptake mechanism and thus , for vaccinia entry . To take an unbiased systematic approach toward the identification of these cellular factors , we developed a system using the model organism Drosophila to perform a high-throughput RNA interference ( RNAi ) screen for cellular kinases and phosphatases required for early steps in vaccinia infection . The Drosophila system is particularly amenable to this approach for a number of reasons including: reduced redundancy in the genome , high conservation with mammalian systems , efficient RNAi , and previous success with this system to identify cellular factors required for viral infection [12] [13] . This Drosophila system is permissive to early steps in the vaccinia infection cycle allowing us to specifically dissect the role of cellular factors involved in the infectious entry process . Using this system , we identified seven genes that contribute to vaccinia infection , including the three subunits of the AMP-activated kinase ( AMPK ) complex , the master energy sensor of the cell . Importantly , the requirement for AMPK in vaccinia infection is conserved in mammalian cells , and is specifically required for vaccinia-induced macropinocytic entry . Further characterization led to the discovery that AMPK controls a variety of virus-independent actin-dependent processes including lamellipodia formation and cell migration . Altogether , we found a new role for AMPK in actin dynamics .
Since vaccinia infection of Drosophila cells has not been reported , we first characterized the course of infection in Drosophila cells . Using a reporter virus expressing Beta-galactosidase ( B-gal ) under the control of an early/late promoter which is active during all stages of vaccinia infection , we found that vaccinia infection is dose-dependent with maximal expression at 48 hours post infection ( hpi ) ( Figure S1 ) . Next , we infected Drosophila cells using reporter viruses that express B-gal under the control of temporally regulated vaccinia promoters that are active during different phases of the virus replication cycle . We found that Drosophila cells were efficiently infected as measured by the production of B-gal from an early/late promoter ( p 7 . 5 ) or by the production of E3L protein , a vaccinia gene product expressed early in infection , while there was very little expression of B-gal from either an intermediate promoter ( G8R ) , or a late promoter ( p11 ) ( Figure 1A ) . Consistent with these findings , we have been unable to detect vaccinia DNA replication ( data not shown ) suggesting a block to infection following early protein synthesis . These findings demonstrate that while vaccinia is unable to complete all stages of the lifecycle in Drosophila cells , entry and early expression occur , providing a model system to study the host factor requirements of vaccinia entry . Previous studies have shown that efficient vaccinia entry is dependent upon the endocytic route of macropinocytosis [4] . In order to assess whether host requirements for vaccinia entry were conserved between Drosophila and mammalian cell lines , we tested whether inhibition of macropinocytosis attenuated infection in these disparate cell types . To this end , we treated cells with several known inhibitors of macropinocytosis and vaccinia entry including; an actin inhibitor Latrunculin A , phosphoinositide-3-kinase ( PI3K ) inhibitor Wortmannin , Na/H antiporter inhibitor EIPA , and PKC inhibitor Rottlerin [4] [7] . We found that each of these drugs significantly inhibited vaccinia infection in both human and Drosophila cells ( Figure 1B–C , quantified in Figure S2 ) . These data show that at least early steps in the viral lifecycle are dependent upon similar pathways in insect cells allowing us to use this model to identify additional factors required for vaccinia infection in mammalian cells . In order to systematically probe the requirements for cellular signaling factors in early vaccinia infection , we developed a quantitative assay amenable to RNAi using virally encoded B-gal expression as a measure of early infection . While a non-targeting negative control dsRNA ( GFP ) had no effect on the percentage of infected cells , knock-down of B-gal by RNAi reduced the percentage of B-gal expressing cells 17-fold , indicating that vaccinia infection can be quantitatively assayed using this system ( Figure 2A ) . Moreover , dsRNA targeting the cellular gene Rab5 , a small GTPase required for many endocytic processes including macropinocytosis [8] also significantly decreased vaccinia infection , validating that we can identify cell-encoded factors required for vaccinia infection using this approach ( Figure 2A and 2D ) . We used this assay to perform an RNAi screen against the Drosophila kinome to identify novel signaling factors that promote vaccinia infection ( Schematic diagram Figure 2B ) . This screen consisted of approximately 440 unique genes ( ∼200 kinases , ∼90 phosphatases , and ∼150 regulator factors ) arrayed onto 384 well plates ( Table S1 ) . Additionally , negative control wells were included containing either no dsRNA ( 15 wells ) or dsRNA targeting GFP ( 28 wells ) , which is not expressed in this system . Lastly , 21 positive control wells with dsRNA targeting lacZ were included ( Figure 2B light blue ) . Drosophila cells were seeded in these pre-arrayed 384 well plates , incubated for three days to allow knock down of each targeted gene , and then infected with vaccinia virus for 48 hours . For the screen , a baseline infection of 10% was within the linear range and was achieved at a multiplicity of infection ( MOI ) of 1 . 25 ( Figure S1C ) . The plates were fixed and processed for immunofluorescence using B-gal expression as a measure of infection , and counter-stained to monitor cell number . Automated microscopy and image analysis were used to quantify the percent infection ( B-gal+/Total Nuclei ) that was transformed into Robust Z scores for each plate , and the Robust Z scores of the 2 replicates were plotted against each other ( Figure 2B ) . Positive candidates were defined as having a Robust Z score of <−2 in duplicate screens ( p<0 . 05 ) . Using these metrics , we identified 8 genes ( 2% , orange and pink Figure 2B ) . In addition to these 8 factors , we identified 20 out of 21 ( 95% ) of the positive control lacZ dsRNAs ( Figure 2B light blue and Table S1 ) and none of the 43 negative controls ( non-targeting dsRNA and empty wells ) . We also monitored the toxicity of the dsRNA treatments and found that none of the 8 genes that inhibited infection were cytotoxic ( <25% decrease in cell number , Table 1 ) . In contrast , we found that while 17 wells reduced cell number by >25% in duplicate screens , none of these genes also inhibited infection . Therefore , our screen revealed host factors required for robust infection that are not required for cell viability . Notably all 8 genes have human homologs ( Table 1 ) , including all 3 subunits of the AMP-activated kinase ( AMPK ) complex ( SNF1A ( AMPKα ) , SNF4Agamma ( AMPKγ ) and alicorn ( AMPKβ ) ) ( Figure 2B pink ) , a heterotrimeric complex involved in maintaining cellular energy homeostasis . RNAi resulted in ∼3-fold reduction in vaccinia infection when AMPK was depleted ( Figure 2C ) . Non-overlapping secondary dsRNAs were generated for each of the 8 genes and were tested to confirm the role of each of these genes in vaccinia infection . Seven of the genes validated ( 88% ) . Importantly , among the genes that validated were those encoding the three subunits of the AMPK complex ( Figure 2D , Figure S3 and Table 1 ) . Moreover , RT-PCR confirmed that AMPKα was depleted by dsRNA treatment against AMPKα ( Figure S4A ) . Furthermore , we found that loss of AMPKα or AMPKγ also led to a defect in early viral mRNA accumulation compared to control ( Figure 2E ) , suggesting that AMPK is required upstream of viral mRNA production in Drosophila , perhaps at the stage of entry . AMPK is an important sensor of intracellular energy that is conserved in eukaryotes ranging from yeast to humans [14] . While Drosophila encodes only one copy of each AMPK subunit , mammals have multiple isoforms of each subunit encoded by several distinct genes ( α1 , α2 , β1 , β2 , γ1 , γ2 , γ3 ) which can produce at least 12 possible heterotrimeric combinations [15] . The lack of redundancy in Drosophila allowed our identification of AMPK by single gene RNAi . We used this Drosophila system as a tool to identify novel host factors that contribute to vaccinia infection , but since Drosophila is not a natural host , we were interested in determining the role of AMPK in a more biologically relevant context . To investigate the role of AMPK in vaccinia virus infection of mammalian cells , we took advantage of mouse embryonic fibroblasts ( MEFs ) that are genetically altered and null for both AMPKα subunits , AMPKα1 and AMPKα2 ( AMPKα1/AMPKα2 −/− ) and verified the lack of these proteins by immunoblot analyses ( Figure S5 ) [16] [17] [18] . These cell lines divide and grow indistinguishably from their sibling control AMPKα1/AMPKα2 +/+ cells ( wild type ) ( data not shown ) . We challenged either the AMPKα1/AMPKα2−/− cells or their sibling control wild type cells with vaccinia virus and measured infection using a plaque assay . This revealed a 20-fold decrease in plaque number and a 15-fold decrease in plaque area in AMPKα1/AMPKα2 −/− compared to wild type cells ( Figure 3A–C ) . The requirement for AMPK was also observed for the closely related poxvirus , cowpox virus ( Figure 3D , quantified in Figure S6 ) . These decreases in infectivity were specific for poxviruses and not simply due to a decrease in overall cell health since several unrelated RNA viruses , including Vesicular Stomatitis virus ( VSV ) grew as well in the AMPK deficient MEFs compared to wild type ( Figure S7 and data not shown ) . This suggests that AMPK deficient cells are capable of supporting all stages of virus infection; including at least some forms of endocytosis and endosomal trafficking since VSV enters cells through clathrin-mediated endocytosis [19] . Therefore , we identified a specific requirement for AMPK in poxvirus infection but not for viral infection generally . In addition to plaque assays we also monitored vaccinia infection by immunofluorescence , immunoblot , and Northern blot and found that there was a significant decrease in vaccinia virus replication in AMPKα1/AMPKα2 −/− cells in each assay ( Figure S8A–C ) . This suggests that AMPK promotes early steps of the vaccinia lifecycle in mammalian cells as well as Drosophila . Finally , to verify that the requirement for AMPK was not MEF-specific we tested whether inhibition of AMPK in the human osteosarcoma cell line U2OS attenuated vaccinia infection using two approaches . First , we pre-treated U2OS cells with the AMPK inhibitor Compound C or vehicle and challenged these cells with vaccinia [20] . Again , we found that inhibition of AMPK attenuated infection ( Figure 4A , B ) . Next , we depleted AMPKα1 , AMPKα2 , or both AMPKα1 and AMPKα2 using siRNAs and observed a decrease in vaccinia infection ( Figure S9A , B , Text S1 ) . We confirmed knock-down by immunoblot ( Figure S9C ) . Together , these data show that vaccinia infection is dependent upon AMPK for infection across disparate cell types including Drosophila , human and mouse . AMPK is activated through phosphorylation of a threonine residue on the catalytic α subunit , which can be triggered by a variety of stimuli including an increase in the cellular ratio of AMP/ATP [21] [22] [23] [24] [25] [26] . Since AMPK promotes vaccinia infection , we tested whether infection activates AMPK . We used a phospho-specific antibody against AMPKα to measure AMPK activation . Treatment with 2-deoxyglucose ( 2DG ) , a known activator of AMPK , led to an increase in AMPK phosphorylation , while little phosphorylation was detected in untreated controls ( Figure 4C ) . Moreover , we observed an increase in phospho-AMPKα within 10 minutes of vaccinia infection ( Figure 4C ) . This increase was not due to changes in total AMPK levels , and suggests that AMPK becomes activated very early in vaccinia infection . Several upstream kinases have been implicated in AMPK activation under different conditions . The classic activator of AMPK is the tumor suppressor LKB1 , which activates AMPK in response to energy deprivation [27] [28] . In Drosophila , LKB1 is the only described upstream kinase required for AMPK activation and lkb1 mutants phenocopy ampk mutants [29] [30] . In contrast , in mammalian systems , LKB1 is the upstream kinase in response to energy deprivation , while additional upstream kinases , such as calcium/calmodulin-dependent protein kinase kinase beta ( CaMKKβ ) have been implicated in AMPK activation under other conditions [31] [32] . We tested whether LKB1 was required for vaccinia infection using cells that are null for LKB1 [33] and complemented with either vector alone ( LKB1−/−; Vec ) , or an LKB1 cDNA ( LKB1−/−; LKB1 ) ( Figure S10 ) and found that loss of LKB1 had no effect on vaccinia infection in mammalian cells ( Figure 4D ) . Likewise , using RNAi to deplete LKB1 in Drosophila cells , we found that it was dispensable for infection by immunofluorescence ( Figure 4E ) and Northern blot ( Figure 4F ) . RT-PCR analysis validated that LKB1 was indeed knocked down in Drosophila cells ( Figure S4B ) . We also tested whether CaMKK , the other well-established AMPK activator in mammalian systems , was required for vaccinia infection . To this end , we pre-treated U2OS cells with the CaMKK inhibitor STO609 prior to infection , and found no effect on vaccinia infection with doses up to 5 µg/ml ( Figure 4A , B ) . Taken together , these data show that vaccinia infection is AMPK-dependent but LKB1 and CaMKK-independent . Given that loss of AMPK led to a decrease in both viral mRNA and protein production ( Figure 2 , Figure S8 ) , we tested whether AMPK was required for efficient virus entry . We monitored viral entry into wild type or AMPKα1/AMPKα2 −/− MEFs using a fluorescence-based assay . We prebound virus to the cells , and then allowed infection to proceed for one hour . Incoming virus particles were visualized using an antibody against L1R , a membrane-bound viral surface protein ( Figure 5A ) . Deconvolution of Z stacks was used to visualize vaccinia inside of cells ( Figure 5A , XZ view ) . Quantification revealed a ∼3-fold reduction in the number of AMPK mutant cells that internalized virus ( Figure 5B ) . These data show that AMPK promotes infection at the stage of entry , although we have not ruled out that virus binding could also be affected by lack of AMPK . We were also interested in whether AMPK played a role in vaccinia infection downstream of entry . First , we monitored both early and late gene expression and found that while the percentage of cells expressing an early protein ( E3L ) or a late protein ( L1R ) was reduced in AMPKα1/AMPKα2 −/− MEFs compared to wild type , 100% of the cells that expressed early genes also expressed late genes , indicating no further block to replication ( Figure S11A ) . We also measured the infectivity of virus produced from AMPKα1/AMPKα2 −/− MEFs . We found a ∼3-fold decrease in virus titer produced from AMPKα1/AMPKα2 −/− MEFs compared to wild type ( Figure S11B ) which is similar to the decrease in viral entry ( Figure 5B ) . These data suggest that while fewer AMPK deficient cells produce virus , the vaccinia released from these cells is as infectious as virus produced from wild type cells . Therefore , while AMPK is important for entry , it is dispensable for later steps in the viral lifecycle . Previous studies established that a major entry route for vaccinia is macropinocytosis , which is required for and induced by vaccinia infection [4] . Since we observed a defect in viral entry in the AMPK mutant cells ( Figure 5 ) , and that inhibitors of macropinocytosis attenuated vaccinia infection ( Figure 1 ) , we tested whether AMPK plays a role in vaccinia-induced macropinocytosis . Macropinocytosis can be directly monitored by fluorescently labeled dextran uptake [34] . Neither wild type nor AMPKα1/AMPKα2 −/− MEFs efficiently endocytosed dextran under resting conditions ( Figure 6A ) . However , upon vaccinia infection , macropinocytosis was dramatically induced in wild type MEFs as measured by an increase in dextran uptake into the cell . In contrast to the large number of dextran punctae observed in the infected wild type MEFs , AMPKα1/AMPKα2 −/− MEFs did not efficiently take up the dextran in the presence of virus ( Figure 6A ) . We quantified the level of vaccinia-induced macropinocytosis in the wild type versus AMPK deficient cells and found an approximately five-fold reduction in the percentage of cells undergoing macropinocytosis ( Figure 6B ) . Furthermore , we found a decrease in vaccinia-induced dextran uptake in human U2OS cells pretreated with Compound C compared to vehicle control ( Figure 6C ) . Quantification revealed an approximately five-fold decrease in the percentage of U2OS cells undergoing vaccinia-induced macropinocytosis when AMPK is inhibited ( Figure 6D ) . Together , these data show that virus-induced macropinocytosis is dependent upon AMPK in disparate cell types and hosts . In contrast to this dependence of macropinocytosis on AMPK , there was no defect in transferrin uptake in AMPKα1/AMPKα2 −/− MEFs ( Figure S12 , Text S1 ) , indicating that receptor-mediated endocytosis is not controlled by AMPK . This is consistent with our observation that VSV infection is not attenuated in AMPKα1/AMPKα2 −/− MEFs as VSV enters cells by receptor-mediated endocytosis and not macropinocytosis ( Figure S7 ) . One of the early steps in macropinocytosis involves extensive actin remodeling characterized by membrane ruffling and lamellipodia formation . We were interested in determining whether AMPK was required for this early step in the macropinocytic pathway , and whether the requirement for AMPK in macropinocytosis was vaccinia-dependent or if AMPK was required more generally for actin remodeling . Therefore , to test whether AMPK controlled actin-dependent remodeling independent of viral infection , we treated cells with phorbol myristic acid ( PMA ) , which induces cells to undergo high levels of actin-mediated ruffling and lamellipodia formation; this is dependent on Rac1 , a small Rho family GTPase that is also required for macropinocytosis and vaccinia infection [7] [35] [36] [37] . Dramatic lamellipodia formation , seen as thick bands of actin at the cell periphery , were observed in wild type MEFs stimulated with PMA , but abrogated in AMPK-deficient cells ( Figure 7A , arrows ) . We also monitored PMA-induced ruffling using live cell imaging and observed a significant defect in the AMPK mutant MEFs ( Videos S1 , S2 , Text S1 ) . In addition , we monitored the localization of Rac1 during PMA-induced lamellipodia formation in the wild type and AMPK deficient cells and found that both Rac1 re-localization and actin mobilization are defective in AMPK deficient cells ( Figure 7B ) suggesting that the defect is upstream of or parallel to Rac1 activation . While our studies identified AMPK as a critical kinase required for vaccinia infection and actin dynamics , we found that LKB1 was dispensable for infection . This led us to test whether actin-dependent lamellipodia formation and ruffling was also LKB1-independent . We found that there was no defect in PMA-induced lamellipodia and ruffling in LKB1-deficient cells ( Figure 7C , arrows ) . This was not unexpected since HeLa cells , a cell type routinely used for studies on actin dynamics , are mutant for LKB1 , and can still undergo lamellipodia formation , and macropinocytosis [4] [38] [39] [40] . Therefore , we found that the actin remodeling activity of AMPK is independent of LKB1 . Since extensive actin remodeling and Rac1 membrane localization are required for lamellipodia formation , macropinocytosis , and vaccinia infection [7] [35] [36] [37] , we tested whether AMPK was required for another Rac1-dependent actin-dependent process namely in vitro wound healing [41] . For these studies , we created wounds in a confluent monolayer of either wild type or AMPKα1/AMPKα2−/− MEFs by scratching the surface , and monitored wound closure over time . Using this assay we found that there was a significant delay in the migration of AMPK-deficient MEFs into the wound compared to wild type cells ( Figure 7D , E ) . While the wound was completely healed 24 hours after wounding in wild type cells , a sizable gap in the monolayer was still present in AMPK deficient cells , indicating a delay in wound healing and reduced motility . During this motile state , cells undergo a dramatic change in shape , with lamellipodia forming at the leading edge directing cell migration to close the wound . We observed the formation of lamellipodia in the WT MEFs at the edge of the wound while AMPK mutant MEFs did not polarize ( Figure S13 ) . Consistent with our observations that the role for AMPK in actin dynamics is LKB1-independent , we found that wound healing is unaffected by the loss of LKB1 ( Figure S14 ) .
Cell penetration is a critically important step in viral infection , and one that is generally driven by cellular factors and signaling pathways . First , viruses must attach to the cell surface and bind to the viral entry receptor , which often initiates signaling within the cell . Next , viruses must fuse or penetrate the cellular membrane either at the cell surface , or in many cases , within intracellular compartments by taking advantage of the endogenous endocytic machinery . Since there is an array of different endocytic mechanisms , there is great diversity in the strategies used by viruses for entry . Therefore , studying virus entry has increased our understanding not only of viral infection , but also of the normal cellular processes of endocytosis [6] . To further dissect the cellular signaling requirements of vaccinia entry , we developed an unbiased loss-of-function screening platform in Drosophila cells , and identified seven cellular factors required for vaccinia infection . Amongst the genes were all three subunits of AMPK , implicating the entire complex in vaccinia infection . Further studies showed that in addition to its role in Drosophila , AMPK is also required in murine and human cells for poxvirus infection at the stage of entry . Moreover , AMPK becomes rapidly activated upon infection with vaccinia , suggesting that virus-induced signals converge on this complex to facilitate entry . Studies indicate that vaccinia virus can enter cells through multiple routes via as of yet unidentified receptor ( s ) . Studies suggest that virus particles can fuse either at the plasma membrane or from within endosomal compartments , dependent on cell type and virus strain [42] [43] [44] [45] [46] . Importantly , a major endocytic pathway for vaccinia entry has recently been described as macropinocytosis [4] [47] . Our data as well as previous reports support the idea that vaccinia uses macropinocytosis for entry , but not exclusively . We and others have shown partial inhibition of vaccinia infection using a variety of drugs that are well-established inhibitors of macropinocytosis [4] [7] . However , in no case was vaccinia entry completely dependent upon macropinocytosis for infection , demonstrating that the virus can use alternative routes for entry . Nevertheless , macropinocytosis is required for efficient entry across broad cell types suggesting that inhibition of this pathway may attenuate infection sufficiently to allow for immune-mediated clearance of the infection . The process of macropinocytosis drives non-specific uptake of extracellular fluid , large portions of the plasma membrane , as well as large particles . Macropinosomes , unlike coated vesicles , are morphologically heterogeneous , and can vary greatly in size from 0 . 2–10 µm in diameter , sufficient to accommodate the large size of vaccinia virus particles ( ∼0 . 3 µm ) [5] [48] . Classic induction of macropinocytosis by growth factor receptor signaling stimulates ruffles , or sheet-like extensions of the plasma membrane , formed by assembly of actin filaments beneath the plasma membrane that form cups that contract and close to form macropinosomes [5] . This process is driven by signaling events initiated at the plasma membrane and are thought to involve a number of kinase families including phosphatidylinositol 5-kinases ( PI5K ) , PI3K , PKC , serine/threonine kinases , and receptor tyrosine kinases . In addition , as many different inducers of macropinocytosis have been identified , there are likely multiple pathways that converge on the activation of macropinocytosis , adding to the complexity of this cell biological pathway [5] . While several specific kinases , such as PAK1 , and LIM kinase have well described roles in macropinocytosis [49] [50] , there is much that remains unclear . Here , we have found a role for an additional kinase , AMPK , in promoting vaccinia entry through its role in macropinocytosis . We have found that AMPK deficiency attenuates vaccinia infection , concomitant with reduced entry and fluid-phase uptake , supporting an important role for AMPK in vaccinia-induced macropinocytic entry . The process of macropinocytosis involves several steps including extensive actin-mediated membrane ruffling , cup formation , and finally cup closure , which requires the fusion of plasma membranes to close off the macropinosome , followed by fission to separate the macropinosome from the plasma membrane [5] . We discovered that AMPK is required for the formation of lamellipodia and affects the recruitment of Rac1 to the cell periphery , suggesting that the role of AMPK in macropinocytosis lies in the initial rearrangement and reorganization of the actin cytoskeleton . While we have shown that AMPK contributes to actin remodeling during vaccinia-induced macropinocytosis , we also found that AMPK plays a role in other virus-independent remodeling processes including cell migration . During this process , forward movement is driven by the extension of a leading edge protrusion or lamellipodium , followed by contraction at the rear . This protrusive force is generated by localized polymerization of actin mediated by Rac1 [51] . In addition to its role in controlling ruffling upstream of macropinocytosis , we found AMPK also has an essential role in cellular motility and wound healing , demonstrating a broad role in Rac1-dependent actin modulation . Previously , Rac1 has been implicated in nitric oxide production and glucose uptake downstream of AMPK [52] [53] . This role for AMPK and Rac1 in glucose uptake via the translocation of the major insulin-responsive glucose transporter GLUT-4 is quite interesting because this may provide a direct link between AMPK's role in energy homeostasis and the cytoskeleton [54] [55] [56] . While the best understood role of AMPK is its role in metabolism , recent evidence suggests this kinase also has a crucial role in regulating cell structure and polarity through engagement with the actin cytoskeleton . In Drosophila , loss of AMPK leads to defects in mitotic division and epithelial cell polarity accompanied by disruption of the actin cytoskeleton [29] . In some mammalian epithelial cell lines , AMPK activation leads to polarization characterized by the formation of an actin brush-border or tight junction assembly [29] [57] [58] . Additionally , AMPK activation can induce astrocyte stellation , and actin stress fiber disassembly [59] . Finally , studies using Compound C and AMPK activators linked AMPK to macropinocytic uptake of albumin in murine macrophages [60] . Taken together with our new data , this accumulating evidence suggests a broad and conserved role for AMPK in a variety of cellular processes that require actin cytoskeletal rearrangements . The precise signaling events that lead to these AMPK-dependent cytoskeletal changes remain unclear . Several upstream kinases have been shown to activate AMPK under different stimuli . The best studied of these is the tumor suppressor LKB1 which activates AMPK in response to changes in cellular energy levels [27] [28] . Additionally , AMPK can be activated in response to changes in intracellular calcium levels by CaMKKβ , and further evidence suggests that TGFβ-activating kinase ( TAK1 ) may serve as a third upstream activator [31] [32] [61] . Previous studies demonstrated that LKB1 is an important mediator of cell polarity at least in part through signaling to AMPK , and has been shown to drive actin brush border formation , and the translocation of apical and basal markers during the establishment of polarity [29] [30] [57] [58] [62] . We found that at least some AMPK-dependent cytoskeletal changes are independent of LKB1 and CaMKK including lamellipodia formation , macropinocytosis and wound healing . These different actin-dependent outcomes could be controlled by the unique downstream Rho GTPase family members that may become activated by AMPK depending on the stimulus and upstream kinase ( such as Rac1 , which is associated with lamellipodia formation and macropinocytosis ) [10] [11] . Study of the Rho family GTPases activated by AMPK under different stimuli may resolve some of these issues . In addition to the three subunits of AMPK discovered through screening the kinome , we identified four other genes that promote vaccinia infection: Pi3K68D , Fab1 , Stam , and CG9311 , all of which have human homologs . Pi3K68D , Fab1 , and Stam are kinases while CG9311 is the only phosphatase identified in the screen . As kinases are druggable targets , and many known factors involved in macropinocytosis are kinases , we are particularly interested in the role of these kinases in vaccinia infection . Both Pi3K68D ( PIK2C2A ) and Fab1 ( PIP5K3/PIKfyve ) have roles in metabolism of phosphatidylinositol ( PtdIns ) , an important component of membrane trafficking , cytoskeletal rearrangements , and macropinocytosis . In particular , Pi3K68D is a member of the class II family of PI3Ks that produce PtdIns ( 3 ) P downstream of growth factor stimulation , and can modulate the activity of Rho GTPases such as Rac1 and Cdc42 . Class II PI3Ks have a critical role in lamellipodia formation and in cell migration , localizing to the leading edge of migrating cells [63] [64] . Interestingly , class I PI3K have also been implicated in vaccinia infection , during virus entry , and also later stages of infection [4] [65] [66] . Fab1 , the PtdIns ( 3 ) P 5-kinase that converts PtdIns ( 3 ) P into PtdIns ( 3 , 5 ) P2 , has been implicated in fluid-phase uptake , transport , and endosomal acidification [67] [68] . The third kinase , Stam ( STAM , STAM2 ) is activated by cytokine and growth factor stimulation , and localizes to early endosomes , where it is involved in endosomal sorting [69] [70] . Since these kinases have roles either in trafficking to or from the plasma membrane , or in cytoskeletal rearrangements , and have been implicated in processes related to macropinocytosis , it is quite possible that they also play a direct role vaccinia entry . Perhaps Pi3K68D is involved in promoting cytoskeletal rearrangements that lead to macropinocytosis , while Fab1 and Stam could be involved sequentially in later entry steps leading to membrane fusion once a macropinosome has been internalized . Rearrangements in the actin cytoskeleton are crucial not only for vaccinia infection , but also for many essential cellular processes including: cell division , establishment and maintenance of polarity , cellular motility , and uptake of extracellular fluids , each of which must be carefully regulated . While these various processes have different outcomes for the cell , they share several important signaling components , with AMPK as a central mediator . Further characterization of AMPK as well as these additional new factors is required to determine their precise role in vaccinia infection and whether they interact with AMPK , macropinocytosis , or other actin-dependent processes . How AMPK activation in response to different signals leads to these disparate changes in the actin cytoskeleton , and how these processes fit into the larger network of AMPK-dependent pathways will drive future studies . Lastly , the development of more selective AMPK inhibitors or other inhibitors of macropinocytosis may be useful against poxviruses , and other viruses that hijack this endocytic route for their entry mechanism .
Drosophila DL1 cells were grown and maintained at 25°C in Schneiders Drosophila media supplemented with 10% FBS ( JRH ) as described [71] . MEFs , BSC-1 and U2OS cells were maintained at 37°C in DMEM supplemented with 10% FBS ( Sigma ) and 10 mM Hepes . HeLa S3 suspension cells were maintained in MEM supplemented with 10% FBS and 0 . 05% Pluronic . BSC-1 cells were maintained in MEM supplemented with 10% cosmic calf serum ( Hyclone ) . All media were additionally supplemented with 100 µg/ml penicillin/streptomycin and 2 mM L-glutamine . LKB1−/− MEFs were complemented with MIGR ( Vector ) or FLAG-LKB1-MIGR ( LKB1 cDNA ) and maintained as above . Vaccinia strains vPRA13 , vSC8 , and vP30CP77 , were grown in HeLa S3 suspension cells supplemented with 2 . 5% FBS , and tittered on BSC-1 cells as described [72] [73] [74] [75] . Cowpox Brighton Red and Vesicular Stomatitis virus ( Indiana ) were used . Antibodies were obtained from the following sources: anti-Bgal ( Promega and Cappel ) , anti-E3L ( gift from S . Isaacs ) [76] , anti-L1R ( R180 gift from G . Cohen and R . Eisenberg ) , anti-Rac1 ( Millipore ) , and anti-AMPK ( Cell Signaling Technology ) . Fluorescently labeled secondary antibodies along with anti-sheep HRP were obtained from Jackson Immunochemicals or Invitrogen . All other HRP-conjugated antibodies were obtained from Amersham . AlexaFluor 488 and 594 phalloidin , FITC-conjugated dextran , and 594-conjugated Transferrin were purchased from Invitrogen . Compound C [20] and STO-609 [77] were obtained from Calbiochem . Additional chemicals were obtained from Sigma . A mini library of dsRNAs generated against Drosophila kinase and phosphatase genes was obtained from N . Perrimon , and aliquoted onto 384 well plates at 250 ng dsRNA/384 well [78] . Secondary amplicons and control dsRNA were designed using SnapDragon and DRSC resources ( www . flyrnai . org ) , and generated as described [79] . For 384 well assays , 16 , 000 DL1 cells were seeded onto 250 ng dsRNA in 10 µl serum free media . One hour later 20 µl complete media was added , and cells were incubated in a humid chamber for 3 days . For other experiments , 2 , 000 , 000 cells were seeded onto 4 µg of dsRNA/6 well in 1 mL serum free media . One hour later 2 mL complete media was added . For viral infections , vaccinia was tittered on BSC-1 cells , and MOIs added to all cell types are based on pfu/ml measured on BSC-1 cells . Media was removed and virus was added in 2% serum medium and incubated at 25°C for Drosophila cells , and 37°C for mammalian cells . Viral innocula used was adjusted to achieve ∼10% infection of Drosophila cells in the primary screen , and ∼20% infection in secondary analysis . Level of infection of mammalian cells varied depending on the assay format . Cells were processed at the indicated time point post infection . Cells were fixed and processed for immunofluorescence as previously described at 48 hours post infection for Drosophila cells and 8 hours post infection in mammalian cells [12] . Briefly , cells were fixed in 4% formaldehyde/phosphate buffered saline ( PBS ) , washed twice in PBS/0 . 1% TritonX-100 ( PBST ) , and blocked in 2% BSA/PBST . Anti-E3L and anti-B-gal primary antibodies were diluted in block , added to cells , and incubated overnight at 4°C . Cells were washed three times in PBST , and incubated in secondary antibody for one hour at room temperature . Cells were counterstained with Hoescht33342 ( Sigma ) . Plates were imaged at 20X for Drosophila cells and 10X for mammalian cells , capturing three images per well per wavelength using an automated microscope ( ImageXpress Micro ) , and quantification was performed using MetaXpress image analysis software . Significance was determined using a Student T-test . Image analysis was used to generate metrics from the captured images including the number of nuclei and the number of infected cells per site . The percent infection was calculated for each site , log-transformed , and the interquartile range ( IQR ) was used to calculate a robust Z score for each site using the following equation: log10 [ ( %infection-median ) / ( IQR*0 . 74 ) ] [80] . Candidates were identified as positive if the average robust Z score of all sites in a well was <−2 in two independent replicates . For protein analysis , MEFs were prechilled to 16°C for 10 minutes and then treated with vaccinia ( MOI 20 ) for 1 hour at 16°C to synchronize infection . Cells were incubated at 37°C for 10 or 30 minutes or treated with 10 µM 2DG for 30 minutes . Cells were then washed briefly in cold PBS and lysed in NP40 lysis buffer supplemented with protease ( Boehringer ) and phosphatase ( Sigma ) inhibitor cocktails . Samples were separated by SDS-PAGE and blotted as described [81] . HRP-conjugated secondary antibodies and Western Lightening Chemiluminescence Reagent were used for visualization . For RNA analysis , cells were lysed in Trizol buffer , and RNA was purified and blotted as previously described with the indicated probes [12] . RT-PCR was performed using M-MLV reverse transcriptase on random primed total RNA ( Invitrogen ) . One µL of the cDNA or a 1∶10 dilution was used for PCR amplification . Viruses were plaqued on MEF or BSC-1 cells as indicated . Confluent monolayers were treated with serial dilutions of virus for two hours , after which the cells were overlayed with agarose followed by crystal violet staining . Plaque number was determined manually , and plaque diameter was measured using MetaXpress software and used to calculate areas . MEFs plated on cover slips were chilled to 16°C for 10 minutes and then treated with vaccinia ( MOI 100 ) at 16°C for 1 hour . Unbound virus was removed , and cells were incubated at 37°C for 1 hour , washed three times in cold PBS , and fixed . Cells were washed in ammonium chloride ( 50 mM ) and PBST and were stained with anti-L1R and then washed and incubated with secondary antibody , Hoescht 33342 , and phalloidin 488 . Cover slips were mounted and imaged using a 63X objective with a Leica DMI 4000 B fluorescent microscope . Images were taken as 0 . 2 um Z-stacks that were deconvolved using AutoQuant X2 software using Adaptive PSF with 20 iterations . Images are displayed as max projections . To quantify , images were randomized and blindly quantified for virus entry ( n>30 for each condition ) . MEFs grown on glass cover slips were chilled to 16°C for 10 minutes and then treated with vaccinia ( MOI 200 ) at 16°C for 1 hour . Unbound virus was removed , and FITC-dextran ( 70 kD , lysine fixable ) was added at 0 . 5 mg/ml . Cells were incubated at 37°C for 20 minutes , washed twice in PBS , and once in pH 5 . 5 buffer ( 0 . 1 M sodium acetate , 0 . 05 M NaCl ) for 5 minutes . Cells were fixed and stained with Hoescht 33342 and phalloidin 594 . Cover slips were mounted and imaged using a 63X objective with a Leica DMI 4000 B fluorescent microscope . Images were randomized and blindly quantified for the percentage of cells undergoing macropinocytosis as defined by >20 punctae per cell . U2OS cells grown on glass cover slips were pretreated with 10 µM Compound C or vehicle for 1 hour and then assayed as above . Cells were grown on glass cover slips and treated with vehicle or 1 µM PMA for 3 hours . For Rac1 localization experiments , cells were blocked in 8% BSA/PBST for 1 hour . Anti-Rac1 ( Millipore ) was added in 1% BSA/PBST overnight at 4°C . Cells were washed 3 times in PBST , and secondary antibodies were added for 1 hour at room temperature . For all experiments , cells were stained with Hoescht 33342 and phalloidin 488 . Cover slips were mounted and imaged using a 63X objective with a Leica DMI 4000 B fluorescent microscope . Cells grown on glass cover slips were chilled to 16°C for 10 minutes and then treated with vaccinia ( MOI 100 ) at 16°C for 1 hour . Unbound virus was removed , and 594-transferrin was added at 20 µg/ml . Cells were incubated at 37°C for 20 minutes , washed twice in PBS , and once in 0 . 1 M sodium acetate , 0 . 05 M NaCl , pH 5 . 5 buffer for 5 minutes . Cells were fixed and stained with Hoescht 33342 and phalloidin 488 . Cover slips were mounted and imaged using a 63X objective with a Leica DMI 4000 B fluorescent microscope . Cells were grown to 100% confluence overnight , then scratched with a pipet tip to wound . Several marks were made along the length of the wound , and were imaged over time , using these marks as guides . Images were analyzed for wound length at the same position over time using MetaXpress software . | Entry is a vital step in establishing viral infection , providing a potential therapeutic target . Many viruses co-op one of the various cellular endocytic routes for entry , making the host factors that contribute to these processes essential for efficient infection . In particular , vaccinia , the prototypical poxvirus , takes advantage of macropinocytosis for efficient uptake . To identify the signaling factors required for entry of vaccinia virus we performed an RNAi screen of the Drosophila kinome for those kinases and phosphatases that facilitate infection . We identified seven genes including the three subunits of AMPK as promoting infection , and found that AMPK was also required in mammalian cells for vaccinia infection . Furthermore , we found that AMPK facilitates vaccinia entry thru its ability to modulate the actin cytoskeleton and macropinocytosis . In addition to promoting vaccinia uptake , we found that AMPK also contributes to other virus-independent but actin-dependent processes including lamellipodia formation and cellular motility , indicating a broad cellular role in regulating actin dynamics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/microbial",
"growth",
"and",
"development",
"infectious",
"diseases/viral",
"infections",
"virology/host",
"invasion",
"and",
"cell",
"entry",
"genetics",
"and",
"genomics/functional",
"genomics"
] | 2010 | A Kinome RNAi Screen Identified AMPK as Promoting Poxvirus Entry through the Control of Actin Dynamics |
Malaria symptoms occur during Plasmodium falciparum development into red blood cells . During this process , the parasites make substantial modifications to the host cell in order to facilitate nutrient uptake and aid in parasite metabolism . One significant alteration that is required for parasite development is the establishment of an anion channel , as part of the establishment of New Permeation Pathways ( NPPs ) in the red blood cell plasma membrane , and we have shown previously that one channel can be activated in uninfected cells by exogenous protein kinase A . Here , we present evidence that in P . falciparum-infected red blood cells , a cAMP pathway modulates anion conductance of the erythrocyte membrane . In patch-clamp experiments on infected erythrocytes , addition of recombinant PfPKA-R to the pipette in vitro , or overexpression of PfPKA-R in transgenic parasites lead to down-regulation of anion conductance . Moreover , this overexpressing PfPKA-R strain has a growth defect that can be restored by increasing the levels of intracellular cAMP . Our data demonstrate that the anion channel is indeed regulated by a cAMP-dependent pathway in P . falciparum-infected red blood cells . The discovery of a parasite regulatory pathway responsible for modulating anion channel activity in the membranes of P . falciparum-infected red blood cells represents an important insight into how parasites modify host cell permeation pathways . These findings may also provide an avenue for the development of new intervention strategies targeting this important anion channel and its regulation .
Plasmodium falciparum is the species responsible for the lethal form of malaria [1] . The disease is caused by the developmental cycle of the blood stage of the parasite [2] . Their proliferation within the red blood cells ( RBCs ) requires amino acids from the degradation of haemoglobin , but it also depends on the uptake of essential nutrients from blood plasma such as panthotenate or glutamate , whereas waste products such as lactate have to be extruded . These compounds are imported/exported via constitutively active endogenous transporters and new parasite-induced permeation pathways ( NPPs ) [3] . NPPs appear about 12–15 h after red blood cell invasion by merozoites and increase noticeably reaching a plateau 36 h post-invasion [4] . The substrate specificities of NPPs have been extensively characterized by means of tracer fluxes and iso-osmotic haemolysis experiments and show permeability for monosaccharide sugars and other small polyols , amino acids , peptides , nucleosides , various monocarboxylates , small quaternary ammonium compounds , and monovalent inorganic anions and cations including Ca2+ ( for review [5 , 6] ) . Furthermore , they have been found to be non-saturable , to have an activation energy lower than that typical of carrier-mediated transports and to be blocked by numerous anion transport inhibitors ( e . g . Furosemide and 5-nitro-2 ( 3-phenylpropylamino ) benzoic acid ( NPPB ) ) leading to the conclusion that NPPs have some of the characteristics of anion channels [6] . It has been previously shown that after invasion , at the trophozoite stage , anion conductances are activated in the red cell membrane [7–9] . Even if the number and the origin ( human or parasite encoded ) of the channels responsible for this increase in membrane conductance are still under debate [10–12] , today there appears to be a consensus that membrane conductance of infected cells is predominantly inwardly rectified and due to the activity of anion channels [11] . Moreover , recent data showed that this inwardly rectified membrane conductance can be accounted for by the activity of two different types of channels [13] . However , whether or not these channels are indeed correlates of NPPs already described by fluxes and semi-quantitative haemolysis experiments has not been established [11] . Nevertheless , in a recent attempt to unify the different data provided by flux experiments , semi-quantitive haemolysis experiments and results reported using the whole-cell configuration of the patch-clamp technique , it was shown that solute transport via NPPs is not consistent with a single channel model [10 , 11] . For example , sorbitol does not enter infected RBCs via a channel with a low open-state probability at positive membrane potential . We have previously shown using cell-attached and inside-out configurations of the patch-clamp technique that one of the anion channels described at the single channel level could be activated in uninfected erythrocytes by stimulation of the cAMP-dependent signalling pathway , while in infected red blood cells this anion channel is activated by the parasite [8 , 14] . Moreover , this increase in membrane conductance following parasite infection can be inhibited by intracellular dephosphorylation [15 , 16] . The cAMP-dependent signalling pathway has attracted interest from a number of groups [17–19] and cAMP-dependent protein kinase catalytic subunit ( PKA-C ) homologues have been reported in P . yoelii [20] and P . falciparum [21 , 22] . The cAMP pathway plays a central role in many developmental processes in eukaryotes [23–25] . It is generally activated by the binding of a ligand to a membrane receptor , which then causes dissociation of heterotrimeric G-proteins . One of the monomeric G-proteins activates a membrane-bound adenylate cyclase , which causes an increase in intracellular cAMP concentration . In un-stimulated cells , PKA-C is bound to an inhibitory regulatory subunit ( PKA-R ) in a tetrameric complex composed of two PKA-R and two PKA-C molecules . Binding of cAMP to the regulatory subunits , each of which contains two cAMP-binding sites , releases the PKA-C subunits , which results in their activation . PKA-C substrates include transcription factors and other proteins involved in developmental processes [17] . PKA activity has been detected in P . falciparum asexual blood-stage parasites and is higher in mature schizonts than in younger stages of parasite development [22] . Treatment of infected cultures with H89 , a chemical inhibitor of PKA , inhibits parasite proliferation [22] . It is unclear however , whether the unique target is the PKA of the parasite , that of the host red blood cell , or another enzyme altogether ( H89 inhibits other protein kinases in addition to PKA ) [26] . Some residues of mammalian PKA-C that are important for binding to the regulatory subunit are not conserved in the PKA-C subunits of P . yoelii and P . falciparum . Biochemical evidence for the presence of cAMP-binding proteins in parasite extracts , which may include parasite PKA regulatory subunits , has been reported [27] . Moreover , cAMP has also been implicated in regulation of the parasite's cell cycle [25] . In addition to PfPKA-C , which is the only PKA-C subunit among all the P . falciparum kinases ( commonly called the P . falciparum kinome ) [28] , nucleotide cyclases [29] and phosphodiesterases [30] have also been described in P . falciparum . To complete the picture of the cAMP pathway in the parasite , we characterised the single PKA-R subunit of P . falciparum ( PfPKA-R ) , and demonstrate that this protein interacts with a cAMP-inducible kinase activity in P . falciparum extracts . Finally , we show that exogenous recombinant PfPKA-R , as well as overexpression of PfPKA-R in transgenic parasites , affects electrophysiological properties of the anion membrane conductance of infected erythrocytes .
BLASTP analyses of predicted P . falciparum proteins in the PlasmoDB database using PKA-R subunits from various eukaryotes as queries identified a single P . falciparum polypeptide displaying significant homology to the query sequences , in agreement with our previous study [28] . The predicted 441-residue protein , which we call PfPKA-R ( PlasmoDB identifier PFL1110c , or GenBank [http://www . ncbi . nlm . nih . gov/]/EBI Data Bank [http://www . ebi . ac . uk/embl/] accession number AJ441326 ) , displays 42 . 7% and 47 . 7% similarity to human IIα and Iα PKA regulatory subunits , respectively ( Figure 1 ) , and contains a Pfam cNMP domain ( PF00027 ) that encompasses two pairs of the PROSITE cNMP-binding motifs ( PS00888 and PS00889 ) . More specifically , PfPKA-R contains two cyclic nucleotide binding site signatures ( Figure 1 ) , called Phosphate Binding Cassettes , that specifically identify proteins belonging to the R-subunit PKA family [31] . PfPKA-R does not appear to possess a dimerisation domain and shows very weak homology to known PKA-R in the C-terminal part of the protein . Overall , P . falciparum PKA-R is most similar to the human RI subunit , but by the presence of a phosphorylatable residue ( serine ) in the hinge region ( K146RLSV150 ) , PfPKA-R also resembles the RII subunit found in higher eukaryotes [32] ( see Figure S1 ) . Thus , the single PKA-R subunit present in P . falciparum shares characteristics with both mammalian RI and RII subunits ( Figure 1 ) . Sequence analysis of the pfpka-r coding region amplified from a cDNA library confirmed the annotation gene prediction proposed in PlasmoDB ( http://www . PlasmoDB . org ) . Northern blot analysis detected a 2 . 2 kb pfpka-r transcript ( see below ) , with higher levels in asexual parasites compared to gametocytes . We have previously reported a similar pattern for pfpka-c mRNA [22] . To determine whether PfPKA-R is able to modulate a cAMP-dependent kinase activity , PfPKA-R was expressed as a Maltose-Binding Protein ( MBP ) fusion protein and incubated with parasite extracts in the presence or absence of cAMP . As a control we incubated the MBP moiety alone with parasite extracts . A kinase reaction was then performed and 32P incorporation into Kemptide was monitored ( Figure 2 ) . PKA activity was normalised to Kemptide-kinase activity in absence of cAMP . There is a significant difference in activity when the kinase assays are performed in presence compared to the absence of cAMP in parasite extracts ( columns 1 and 2 , p = 0 . 0002 , n = 2 ) , as we previously reported [22] , and compared to parasite extracts incubated with only the MBP moiety ( columns 5 and 6 , p = 0 . 0032 , n = 2 ) showing that MBP has no significant effect on PKA activity on its own . In contrast , addition of MBP-PfPKA-R causes a 55% decrease in the cAMP-dependent Kemptide kinase activity present in the parasite extract ( columns 3 and 4 , p = 0 . 1524 , n = 2 ) , suggesting in this case that the cAMP interacts with the endogenous PfPKA-R and with the recombinant MBP-PfPKA-R . The whole-cell configuration of the patch-clamp technique allows one to follow changes in membrane conductance after erythrocyte infection . Although this configuration gives activity measurements of all channels present in the red blood cell membrane , it is limited experimentally , as one cannot change the cytosolic medium during the time course of recordings . For this reason , in this study non-diffusible compounds were added to the pipette solution and the effects were compared to control cells tested the same day in parallel . An inwardly rectified whole cell current is inducible in uninfected erythrocytes by addition of bovine PKA and ATP , and the resulting membrane currents are similar in shape to those measured in P . falciparum-infected red blood cells [8] . The addition of alkaline phosphatase ( 0 . 1 u/ml ) results in a complete inhibition of current , suggesting regulation by phosphorylation ( Figure 3 ) . Thus , in order to test the implication of a cAMP-dependent kinase , we added the natural inhibitor of mammalian PKA , PKI ( 1 μg/ml ) , into the pipette . PKI is a small heat-stable protein that is highly specific ( Ki = 2 . 3 nM ) [33] interacting with the mammalian catalytic subunit within the cleft ( between the two lobes ) and with the larger lobe of the enzyme [34] . Maximum inhibition of current is reached after 10–15 minutes and corresponds to 59 . 2 ± 10 . 0% of membrane currents inhibition ( n = 4 ) . As expected , according to the mode of inhibition of PKA by PKI in mammals , this inhibition could not be reversed by competition with exogenous 5 mM cAMP , suggesting a direct interaction between the PKI and the PKA-C target ( Figure 3A ) . We next investigated whether or not addition of recombinant PfPKA-R into the pipette could repress the P . falciparum-induced membrane currents . Addition of MBP-PfPKA-R down-regulates whole-cell membrane conductance in a time-dependent manner ( Figure 3C ) . Maximum inhibition of membrane current is 43 . 0 ± 7 . 0% ( n = 4 ) , whereas MBP alone did not change the shape of currents , nor membrane conductance during at least 30 minutes ( Figure 3B ) . Moreover , the down-regulation appears to be regulated by cAMP , as addition of 5 mM cAMP resulted in <10% inhibition ( versus >40% in the absence of cAMP , see Figure 3A ) . These results demonstrate that addition of exogenous PfPKA-R interferes with the parasite-dependent activation of infected red blood cell membrane conductance . To investigate PfPKA-R function in vivo , we generated a P . falciparum line that overexpresses the protein from the pHL-dhfr-pfpka-r vector . Using parasites transfected with pHL-dhfr-luciferase as a control , we assessed the expression of the pfpka-r transgene . The expression occurs in asexual stages , as observed in wild type parasites ( Figure 4A ) , but was high and constant throughout the erythrocytic cycle ( Figure 4C ) , as expected for a transgene driven by the hrp3 promoter [35] . In wild-type and in pHL-dhfr-luciferase-transfected control parasites , expression of pfpka-c and pfpka-r peaks at the schizont stage , with very low levels in earlier stages of the asexual cycle ( Figure 4B and 4C and http://www . plasmodb . org/ ) . In contrast , the PfPKA-R overexpressers produced large amounts of pfpka-c transcripts at the ring and trophozoite stages ( Figure 4B ) . This might be explained by the parasite's attempt to compensate any deleterious effects stemming from overexpression of the regulatory subunit . Growth of PfPKA-R-overexpressing parasites was monitored for five days . While the growth rate of pHL-dhfr-luciferase transgenic parasites was similar to that of the parental wild type NF54 strain ( p = 0 . 6 , n = 3 after five days , Figure 5A ) , overexpression of PfPKA-R reduced parasite growth rate by 78 ± 3% after five days ( p < 0 . 05 , n = 3 ) ( Figure 5A ) . In order to check if reduced growth was due to dampened cAMP signalling , we monitored the growth of NF54 ( and of NF54 transformed either with pHL-dhfr-pfpka-r , or pHL-dhfr-luciferase ) in the presence of a phosphodiesterase inhibitor ( IBMX ) . Addition of IBMX ( 500 μM ) to the culture medium caused a rise in cAMP levels in control infected cells and in parasites overexpressing PfPKA-R from 0 . 20 ± 0 . 03 pmol/108 cells to 0 . 82 ± 0 . 05 pmol/108 cells ( p < 0 . 05 , n = 2 ) and from 0 . 60 ± 0 . 20 pmol/108 cells to 1 . 40 ± 0 . 34 pmol/108 cells ( p < 0 . 05 , n = 2 ) respectively , as previously described [25] . This partially restored the growth rate of parasites overexpressing PfPKA-R , reaching 66% ± 7 ( n = 3 ) of the normalized parasitaemia of non-transformed NF54 parasites ( p > 0 . 1 , n = 3 , Figure 5B ) . However , IBMX stimulation of the pHL-dhfr-luciferase transformed parasites caused a dramatic fall of the growth to 43% ± 6 of the normalized parasitaemia of non-transformed NF54 parasites ( p < 0 . 05 , n = 3 , Figure 5B ) . A similar phenomenon is observed in non-transformed NF54 parasites ( data not shown ) . Moreover , addition of the permeable analogue of cAMP ( 8Br-cAMP , 0 . 1μM ) , enhanced the growth of the PfPKA-R overexpressing parasites reaching 82 ± 19% ( n = 3 ) of the normalized parasitaemia of non-transformed NF54 parasites ( p = 0 . 6 , n = 3 , Figure 5C ) . Stimulation by exogenous 8Br-cAMP of pHL-dhfr-luciferase transformed parasites caused a decrease in growth ( Figure 5C ) that was not significant ( p = 0 . 6 , n = 4 ) , as previously seen with IBMX . Together , these results suggest that the level of intracellular cAMP has to be finely tuned in order to ensure normal cell cycle and differentiation . The PfPKA-R overexpressor and control parasite populations were analysed using the whole-cell patch-clamp method . While the luciferase expressing population showed membrane currents similar to those found in non-transfected cells , the population overexpressing PfPKA-R presented a clear reduction of membrane currents ( 40 . 4 ± 6 . 6% , n = 6 ) ( Figure 5D and data not shown ) . The inhibition was similar in amplitude to that observed when the MBP-PfPKA-R recombinant protein was added to the patch pipette ( see Figure 3A and 3C ) . Interestingly , 24 h stimulation of the overexpressing PfPKA-R cells , in the presence of IBMX , restored the membrane conductances ( Figure S2 ) , reaching values similar to those observed in control cells either for inward currents ( p = 0 . 20 , n = 4 ) , or outward currents ( p = 0 . 15 , n = 4 ) . However , neither stimulation by IBMX during time course of the patch-clamp experiments , nor addition of cAMP into the patch pipette changed membrane conductance of the cells overexpressing PfPKA-R cells ( data not shown ) . In order to test whether PfPKA-R overexpression could interfere with NPP activity , semi-quantitative haemolysis experiments were carried out using sorbitol as permeating substrate through NPPs . In parasites overexpressing PfPKA-R permeability to sorbitol appeared reduced , since t1/2 of lysis was significantly increased ( 18 . 2 ± 1 . 6 min , n = 3 ) compared to what is observed in control cells ( 12 . 7 ± 0 . 8 min , n = 3 , p = 0 . 03 , Figure 6A ) . However , sensitivity to the anion carrier inhibitor NPPB was not altered , since IC50 measurements in our conditions are similar in parasites overexpressing PfPKA-R and in control parasites ( 4 . 5 ± 1 μM and 7 ± 1 μM respectively n = 3 , Figure 6B ) .
In contrast to mammalian cells that have several PKA-C and PKA-R subunits , we show here that cAMP signalling in P . falciparum is extremely reduced , as the parasite possesses only a single regulatory subunit in addition of its single PKA-C subunit . Overexpression of the regulatory subunit in transgenic parasites , or addition of recombinant PfPKA-R or PKI to the pipette in patch-clamp experiments , significantly down-regulates the conductance of one of the erythrocyte plasma membrane anion channels that we have previously described at the single channel level [36] . In this context , it is worth noting that no gene encoding PKI can be detected in the P . falciparum genome , although we have shown previously that P . falciparum PKA-C activity is somewhat PKI sensitive [22] . One of the P . falciparum adenylate cyclases that have been reported has a putative N-terminal ion channel domain encoding a canonical S4 voltage sensor [37] . This leads to the intriguing possibility that cAMP signalling in malaria parasites may regulate anion conductance at the plasma membrane . Membrane currents measured in this study arise from the activity of two different types of chloride channel [13]: besides the activity of the PKA-dependent chloride channel , another type of chloride channel belonging to the ClC-2 family is constitutively active in infected cells . Although the possible activation of this channel type by phosphorylation is still controversial [38 , 39] , this could explain why addition of PfPKA-R failed to completely inhibit anion transport . Transgenic parasites that overexpress the regulatory subunit have a reduced growth ( Figure 5A ) and normal growth rate is largely restored upon enhancement of intracellular cAMP levels ( Figure 5B and 5C ) . This is consistent with our previous observation that H89 , which inhibits PfPKA kinase activity in vitro , blocked parasite growth [22] and the more recent report that the parasite cell cycle is also regulated by cAMP [25] . This could also explain why changes in intracellular cAMP concentrations following addition of IBMX , seems to be lethal for the parasite . During erythrocyte development cAMP levels appear to be finely regulated , as over-stimulation of cAMP production tends to the accumulation of parasites at the schizont stage [25] . Furthermore , IBMX , is poorly selective between cAMP phosphodiesterases and cGMP phosphodiesterases , so we cannot exclude the possibility that the effects observed in the presence of IBMX are due to increases in intracellular concentrations of other cNMPs , as has been shown for exflagellation [40 , 41] , or the growth defect observed in the presence of 8-Bromo-cGMP ( ED50 10 μM ) [30] . P . falciparum codes for four different phosphodiesterases , but only one ( PfPDE1 ) has been characterised and , shows no cAMP hydrolytic activity up to 5 mM cAMP , but has 135 time higher cGMP hydrolysing activity without any competition between cAMP and cGMP [30] . In addition , nor can we rule out the possibility that the effects of cAMP agonists ( IBMX and 8-Br-cAMP ) observed in the present study are a consequence of a rise in intracellular Ca2+ concentrations , as described by Beraldo et al . [42] . Indeed , it is known that modulation of Ca2+ signalling can be modulated by a cAMP-dependent pathway and some evidence suggests an important role for Ca2+ in the control of key processes in Plasmodium such as genes expression and cell cycle progression [25 , 43] . Therefore , it is possible that PfPKA regulates parasite growth via other factors independent of anion channel conductance , as the lack of a similar channel in erythrocytes from cystic fibrosis patients does not seem to interfere with parasite growth [14 , 16] . Moreover , repeated attempts to disrupt the PKA regulatory or catalytic domain in P . berghei , the rodent parasite , failed , suggesting the crucial role of the protein for the malaria parasite ( Tewari , Moon and Billker , pers . com . ) . In patch-clamp experiments on erythrocytes infected by PfPKA-R overexpressing parasites , the addition of non-permeable cAMP in the patch-clamp pipette does not release the inhibition of anion conductance , whereas 24 h IBMX stimulation restore the membrane conductance to the level of controlled cells . This raises the question of the localisation of the PfPKA-R . Previous studies have shown that a PEXEL/HT motif promotes export of proteins from the parasite into the host cell [44 , 45] . PfPKA-R and PfPKA-C sequences do not have a recognisable PEXEL/HT motif , but a parasite antigen ( PfEMP1 ) present in the red blood cell plasma membrane and the kinase PfGSK-3 [46] , which is addressed to the Maurer's cleft also lack recognisable export motifs [47] . cAMP in infected cells is most likely synthesized in the parasite cytosol and cAMP is normally membrane impermeable . However , we cannot rule out that some parasite-derived cAMP might get into the erythrocyte cytosol via the non specific channel that can pass soluble macromolecules of up to 1400 Da and functions as a high capacity , low affinity molecular sieve [48] . We entertain here , the possibility that overexpression of the PfPKA regulatory subunit leads it to act as a sink for cAMP thereby reducing available cAMP levels , both within the parasite and eventually in the erythrocyte cytosol . In this scenario , addition of recombinant MBP-PfPKA-R would limit the release of monomeric active PKA catalytic subunit ( host PKA-C , or secreted PfPKA-C ) , and thus the activation of subsequent cellular mechanisms such as the activation of the anion transport . The delay observed in haemolysis for the parasites overexpressing PfPKA-R , together with and the complete inhibition of membrane currents by unspecific dephosphorylations suggest that some NPP activity is under the dependence of phosphorylation via cAMP-dependent kinases , either directly or indirectly . In the absence of compelling localization data for PfPKA-R and PfPKA-C , we cannot exclude that the effect of PfPKA-R on anion conductance at the erythrocyte membrane may be mediated indirectly , through PKA-dependent regulation of other effectors . In this respect , it is worth mentioning that several protein kinases and phosphatases have been shown to be targeted to the erythrocyte and cAMP-dependent PKA may be involved in their activation and/or secretion [46 , 49 , 50] . It has been known for some time that non-infected erythrocytes have a PKA activity and regulatory RI and RII subunits , with the RI subunit being associated with the plasma membrane [51] . PKA activity at the plasma membrane could contribute to the low-level anion channel conductance of non-parasitized red blood cells [8] . In extracts from P . falciparum-infected erythrocytes host cell PKA-C appears to be cleaved , although the physiological significance of this observation has not been established [22] . Although a role for PfPKA-C in the regulation of anion channels seems likely , we cannot exclude the possibility that host PKA-C contributes to induction of anion conductance in P . falciparum-infected red blood cells . In this scenario , erythrocyte PKA-C would become active at the schizont stage , perhaps following a parasite-dependent increase in cAMP levels in the red blood cell cytosol . At the red blood cell plasma membrane erythrocyte PKA-C is associated with the RI subunit and the recent report that RI can bind integrin α4β1 and recruit PKA-C to phosphorylate the integrin's cytoplasmic tail [52] opens up a wealth of new possibilities for parasite-modulation of integrin function in malaria . Even though erythrocytes do not express α4β1 one possibility that we are currently testing is that the R subunit ( host and/or parasite ) binds directly to another integrin and regulates anion channel conductance in an integrin-dependent fashion , as a role for β1integrin in anion channel conductance has been reported for ventricular myocytes [53] . Clearly , other integrin-dependent signalling pathways in P . falciparum-infected hepatocytes and erythrocytes could be also modulated by PKA , whether it be of erythrocyte or parasite origin .
Primers were designed for amplification and cloning of the entire ORF of 1326 bp ( PlasmoDB identifier PFL1110c ) into the pMALc2X vector ( New England Biolabs ) . The primers contained EcoRI ( forward primer: CCGGAATTCATGGGCAATGTGTGCACATGG ) , or HindIII ( reverse primer: CCCAAGCTTTTAATTTTCATCAATACAA-GTTG ) restriction sites ( single underline ) . After amplification from a P . falciparum cDNA library ( kindly provided by Alister Craig ) with a Taq DNA polymerase ( Takara ) , the PCR product was digested with EcoRI and HindIII prior to insertion in the pMALc2X vector ( New England Biolabs ) . The pMALc2X–pfpka-r plasmid was electroporated into E . coli BL21 , and the insert was verified by sequencing prior to expression of the recombinant protein . Protein expression was induced for 3h at 37°C with 0 . 3 mM isopropyl-α-D-galactoside ( IPTG ) , after the 250 ml culture has reached an OD600 value of 0 . 6 in 2xYT medium with 100 μg/ml ampicillin . All purification steps were performed at 4°C . Bacterial pellets were lysed with lysosyme and by sonication in 5 ml of lysis buffer ( 50 mM Tris pH7 . 5 , 50 mM NaCl , 5 mM EDTA , 1 mM phenylmethylsulfonyl fluoride [PMSF] and ComplexTM protease inhibitors protease inhibitor tablet from Roche Molecular Biochemicals ) . Lysates were cleared by centrifugation ( 11 , 000 rpm , 4°C , 30 min ) and the soluble fraction was incubated for 90 minutes at 4°C under mild agitation with 0 . 5 ml of amylose beads ( New England Biolabs ) . The slurry was washed in lysis buffer and the MBP-PfPKA-R fusion protein was eluted with elution buffer ( 50 mM Tris pH8 . 0 , 50 mM NaCl , 0 . 1 mM EDTA , 1 mM PMSF , ComplexTM protease inhibitors protease inhibitor tablet and 12 mM maltose ) . P . falciparum 3D7 maintained in a modified erythrocyte culture were synchronized by 5% sorbitol . The parasites were released from infected erythrocytes by saponin ( 0 . 1% w/v ) lysis and P . falciparum pellets were sonicated in RIPA buffer ( 30 mM Tris , pH8 . 0 , 150 mM NaCl , 20 mM MgCl2 , 1 mM EDTA , 1 mM dithiothreitol , 10 μM ATP , 0 . 5% Triton X-100 , 1% Nonidet P-40 , 10 mM β-glycerophosphate , 10 mM NaF , 0 . 1 mM sodium orthovanadate , 1 mM phenylmethylsulfonyl fluoride , 10 mM benzamidine , and ComplexTM protease inhibitors ) . The lysates were cleared by centrifugation ( 15 , 000 rpm for 15 min at 4°C ) , and the total amount of proteins in the supernatant was measured by a Bio-Rad protein assay , based on the method of Bradford , using BSA as a standard . PKA activity was assayed by measuring 32P incorporation into a synthetic peptide ( Kemptide L-R-R-A-S-L-G , Promega ) , as described previously [22] . Briefly , in a standard volume ( 30 μl ) containing ( 50 mM MOPS pH7 , 0 . 5 mM MgCl2 , 40 μg/ml BSA , 50 mM NaF , 1 mM β glycerophosphate , 1 mM PMSF and Complex TM mixture tablets Roche Molecular Biochemicals ) , reactions were initiated by addition of 300 μM Kemptide and 15 μM ATP / 5 μCi of ( γ-32P ) -ATP . The reactions were performed in presence or absence of 40 μM cAMP ( prior incubation at 30°C for 15 min ) , and in presence or absence of recombinant MBP-PfPKA-R ( or MBP as a control ) . Significance was assessed using the Fisher F test and Student's t-test . Approximately 5 micrograms of total RNA were electrophoresed in a 1% agarose/formaldehyde gel . After RNA visualisation with ethidium bromide staining , which confirmed that equivalent amounts of RNA were loaded in the two samples , gel was blotted to Hybond-C nitrocellulose ( Amersham ) . Filters were cut , and hybridised with 32P-labelled probes specific for the pfpka-c and the pfpka-r genes , respectively . 735 bp PCR fragment for pfpka-c , and the whole pfpka-r gene ( 1326 bp ) were amplified with the following primers: CATGGATCATTCAAAGATGAC ( PfPKA F2 ) ; GGAAGATCTCTACCAATCATAAAATGGATCATTTTC ( PfPKA BglII stop R ) ; CCGGAATTCATGGGCAATGTGTGCACATGG ( PfPKA SUR EcoRI F ) ; CCCAAGCTTTTAATTTTCATCAATACAAGTTG ( PfPKA SUR HindIII R ) . Hybridisation signals were revealed after autoradiographic exposure ( 40 h ) . Oligonucleotides were designed for amplification containing PstI and HindIII restriction sites ( single underline in primers ) ( forward primer: ATACTGCAGATTATGGGCAATGTGTGCACATGG; reverse primer: ATAAAGCTTAACGACGGCCAGTGCCAAGCT ) . After amplification , PCR products were restriction digested with PstI and HindIII prior to insertion in the expression vector pHL-dhfr [50] . The entire pfpka-r coding region was amplified from cDNA and cloned into the transfection plasmid pHL-dhfr , a derivative of pHLH1 vector [54] . pHL-dhfr-luciferase includes a luciferase open reading frame ( luc ) flanked by the P . falciparum 5′hrp3 promoter and the 3′hrp2 terminator . The vector contains the human dihydrofolate reductase sequence ( dhfr ) , which allows selection of stable transformed parasites NF54 isolate ( from which the 3D7 line is derived ) by addition of 40 ng/ml pyrimethamine in the culture medium . The pHL-dhfr-pfpka-r plasmid differs from the pHL-dhfr-luciferase plasmid by replacement of the luciferase open reading frame by that of pfpka-r . Parasites were transfected , as previously described [55] . Briefly , uninfected human erythrocytes were electroporated with 50 μg of plasmid DNA in complete cytomix medium ( 120 mM KCl ; 0 . 115 mM CaCl2 ; 5 mM MgCl2 ; 5 mM K2HPO4 ; 5 mM KH2PO4 ; 2 mM EGTA and 30 mM Hepes pH7 . 6 adjusted with NaOH ) using a Bio-Rad gene pulser and 0 . 2 cm cuvettes ( conditions 0 . 31 kV and 960μF ) . Erythrocytes from two electroporations were combined for each 5 ml culture and inoculated with late stage parasites purified using a Percoll/sorbitol . Validation of the transfection was done by Q-PCR . After transfection , and the selection by pyrimethamine , more than two months were necessary to recover significant amount of parasites . Moreover , due to the defect of growth of the pfpka-r transformed parasites , the cultures were maintained at the beginning with a 2 . 5% hematocrit instead of the 5% hematocrit generally used , in order to ensure the possibility of adding fresh blood more frequently , improving the fitness of the culture and the relative amount of infected cells . Prior RNA extraction , synchronized infected RBCs ( Rings , trophozoites and schizonts ) were washed 2 times in 1× PBS , permeabilized with 0 . 05% saponin , and then pelleted parasites were rapidly washed 3 times with 1X PBS . RNA extraction was performed with the TRIZOL LS Reagent® as previously described [56] . RNA to be used for cDNA synthesis was treated with Deoxyribonuclease I® ( Invitrogen ) as described by the manufacturer . A total of 1600 ng of RNA was digested in a 20 μl reaction . Samples were incubated at room temperature for 30 minutes followed by 10 minutes heat inactivation at 65°C . cDNA synthesis was performed with Superscript II Rnase H reverse transcriptase ® ( Invitrogen ) with random primers ® ( Invitrogen ) as described by the manufacturer . cDNA was synthesized from 800 ng RNA in a 40 μl reaction . For each cDNA synthesis reaction , a control reaction without reverse transcriptase was performed with identical amounts of template . Gene specific primers were designed for the pfpka-c ( PFI1685w ) and pfpka-r genes using primer 3 software ( http://frodo . wi . mit . edu/cgi-bin/primer3/primer3_www . cgi ) ( forward primer :TTAATGACGACGGTTCAAGC [PfPKAr_qf]; reverse primer : TCCAGTCACCATATGCTTCG [PfPKAr_qr] and forward primer : TGGATGTTTTTATGCAGCTCAG [PfPKAc_qf]; reverse primer TCCATGTCCGACGTTCAATA[Pf©PKAc_qr] ) . With this primer set 153 bp fragments of pfpka-r gene and 223 bp fragments of pfpka-c present in the NF54 Plasmodium falciparum strain are generated . Amplification efficiency was verified by testing all primer pairs using a two 3 log range of concentrations of genomic DNA obtained from transformed parasites and two control genes , namely arginyl-tRNA synthetase ( PFL0900c ) and seryl-tRNA synthetase ( PF07_0073 ) kindly provided by T . J . Templeton and C . Lavazec . Real-time PCR was performed using an ABI Prism 7900HT sequence detector ( Applied Biosystems ) . Reactions were prepared in 25 μl volume using SYBR Green PCR master mix ( Applied Biosystems ) and 1 μM primers . Specificity of amplification was confirmed by melting-curve analysis for each PCR product . Triplicate PCR were analyzed for each sample . The genes transcriptions were expressed relative to the expression of a single copy housekeeping gene ( ΔCt method ) . Briefly , the ΔCt for each individual primer pair was determined by subtracting the measured Ct value from the Ct value of the control seryl-tRNA synthetase ( User bulletin 2 , Applied Biosystems , http://www . appliedbiosystems . com/ ) . ΔCts were then converted to relative copy numbers with the formula 2ΔCt . The effects of drugs on the growth of in vitro parasite cultures were determined as follows . Erythrocyte suspensions with ring-stage synchronized parasites were distributed in triplicate into 24-well plates at 2% parasitemia and 2% hematocrit . The incubations were performed for 72 h at 37°C in a closed chamber with controlled atmosphere in stationary condition . The proportion of different parasite stages as well as parasitemia was determined by analyzing 1 , 000 RBCs on Giemsa-stained slides . As a negative control , cultures treated with solvent only ( DMSO ) were assayed . The total parasitemia was an arithmetic sum of rings , trophozoites , and schizonts at each time point . The whole-cell configuration of the patch-clamp technique was assessed by the development of small capacitance transient and reduction of access resistance . Cation movements across the membrane from the exterior ( bath ) to cytoplasmic side is defined as inward current and shown as downward deflection in whole cell recordings . Seal resistances were 4–20 GΩ . Patch pipettes ( tip resistance 10–20 MΩ ) were prepared from borosilicate glass capillaries ( GC150 TF-10 , Clark Medical Instruments , PHYMEP , France ) pulled and polished on a Werner Zeitz DMZ programmable puller ( Augsburg , Germany ) . The ruptured patch whole-cell configuration was used to record whole-cell currents . Whole-cell currents were recorded using a RK400 ( Biologic , France ) amplifier , with voltage command protocols generated and the currents analyzed using the WCP Software ( WCP V3 . 3 . 3 . Software , Strathclyde , UK ) by evoking a series of test potentials ( VT ) from −100 to +100 mV in 10 mV steps for 500 ms from a holding potential ( VH ) of 0 mV . Data for the construction of I–V curves were the mean current measured between 200 and 400 ms . Data are given as mean values ± S . E . M . Significance was assessed using the Fisher F test and Student's t-test . In all cases , n denotes the number of cells tested . Different bath ( 150 mM NMDG-Cl , 1 . 4 mM CaCl2 , 1 mM MgCl2 , 15 mM Hepes , 10 mM glucose , pH7 . 4 , 320 ± 5 mOsmol , pCa3 ) and pipette ( 150 mM NMDG-Cl , 0 . 29 mM CaCl2 , 1 . 22 mM MgCl2 , 5 mM EGTA , 10 mM Hepes-Tris , 10 mM glucose , pH7 . 2 , 320 ± 5 mOsmol , pCa7 ) solutions were used to ensure proper free calcium concentrations at both sides of the membrane . Since temperature has no influence on whole cell currents in infected red blood cells all experiments were performed at room temperature . RBCs were obtained from healthy volunteers , who gave informed , written consent , in accordance with the Declaration of Helsinki , cAMP ( cyclic adenosine-monophosphate ) , alkaline phosphatase , 8Br-cAMP ( 8-Bromide-cyclic adenosine-monophosphate ) , PKI ( Protein Kinase Inhibitor ) , NMDG Cl ( N-methyl-D-glucamine chloride ) , IBMX ( isobutylmethylxanthine ) and NPPB ( 5-nitro-2- ( 3-phenylpropylamino ) benzoic acid ) were purchased from Sigma ( Saint Quentin Fallavier , France ) . P . falciparum-infected cells were submitted to 24 h of IBMX treatment . Following treatment cells were washed twice in 1X PBS and analysed with the cAMP immunoassay kit ( GE Healthcare , Amersham Biotrak ) following protocol N°3 . For standard semi-quantitative haemolysis assays , haemoglobin release was used to estimate lysis time , as previously described [57] . Culture suspensions ( 2–5% parasitaemia ) were washed three times in culture medium without serum and resuspended at 50% hematocrit . Time courses started with the addition of a 0 . 4 ml aliquot of cell suspension to 3 . 6 ml of the sorbitol iso-osmotic solutions ( 300 mM sorbitol , 10 mM hepes , 5 mM glucose , pH 7 . 4 ) to give a cell concentration of approximately 0 . 5 . 108 cells/ml . Experiments were performed in triplicates . At predetermined intervals ( 0 , 3 , 5 , 10 , 15 , 30 , 60 min ) 0 . 5 ml aliquots of the suspension were transferred to microcentrifuge tubes containing 0 . 5 ml of an ice-cold “stopping solution” ( 400 mM sucrose in H2O ) . The tubes were centrifuged for 30 seconds then 0 . 9 ml of the supernatant solution was transferred to another tube for the subsequent spectrophotometric estimation of haemoglobin concentration by absorption at a wavelength of 540 nm ( A540 ) . In all such experiments the A540 value corresponding to full haemolysis of trophozoite-infected erythrocytes was estimated from the final A540 value achieved in the supernatant solution from infected cells suspended in an iso-osmotic sorbitol . The percent lysis values at different times were fitted by nonlinear regression using equation for a sigmoidal shape according the equation y = a/ ( 1+exp ( − ( x−x0 ) /b ) ) , where y is the percent lysis value , a is the maximal lysis , x is the sampling time , x0 is essentially the t1/2 of lysis , and b represents the variability of cells in the population [58] . The t1/2 of lysis calculated for each experiment was then compared using a paired two-tailed Student's t-test . When drugs were tested , the percentage of inhibition was determined relative to non-treated cells when haemolysis was at maximum . | By replicating within red blood cells malaria parasites are largely hidden from immune recognition , but within mature erythrocytes nutrients are limiting and accumulation of potentially hazardous metabolic end products can rapidly become critical . In order to survive within red blood cells malaria parasites , therefore , alter the permeability of the erythrocyte plasma membrane either by up-regulating existing carriers , or by creating new permeation pathways . Recent electrophysiological studies of Plasmodium-infected erythrocytes have demonstrated that these changes reflect trans-membrane transport through ion channels in the infected erythrocyte plasma membrane . Protein phosphorylation has been documented in protozoan parasites for a number of years and is implicated in key processes of both parasites and parasitized host cells . It has been established that cAMP-dependent regulated pathways are able to activate ion channels in the red cell membrane and a better understanding of how the parasite manipulates cAMP-dependent signaling to activate anion channels could be important in developing novel strategies for future anti-malarial chemotherapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"plasmodium",
"physiology",
"microbiology",
"eukaryotes",
"homo",
"(human)"
] | 2008 | Plasmodium falciparum Regulatory Subunit of cAMP-Dependent PKA and Anion Channel Conductance |
Formerly known as the Malaysian hunter gatherers , the Negrito Orang Asli ( OA ) were heavily dependent on the forest for sustenance and early studies indicated high prevalence of intestinal parasitism . Initiation of a redevelopment program in the 1970s aimed to demarginalize the OA was expected to reduce soil transmitted helminth ( STH ) infections . Gradually , the OA were relocated to new resettlement areas at the peripheries . The aim of this study was to compare STH infections between Negritos who are still living in the inland jungle with those living in resettlements . A total of 416 Negrito participants were grouped into two categories of communities based on location; Inland Jungle Villages ( IJV ) ; and Resettlement Plan Scheme ( RPS ) . Iodine wet mount , formalin-ether sedimentation , modified Trichrome and modified Ziehl-Neelsen staining and Kato-Katz methods were performed on stool samples . A questionnaire was used to collect information regarding demographic , socioeconomic , environmental and hygiene behaviors . Prevalence of STH was significantly higher in IJV ( 91 . 3% ) versus RPS ( 83 . 1% ) ( P = 0 . 02 ) . However , the percentage of individuals with severe intensity of Trichuris trichiura infections was significantly higher in the RPS ( 17 . 2% ) compared to IJV ( 6 . 5% ) ( P = 0 . 01 ) . Severe Ascaris lumbricoides infection was observed at 20 . 0% amongst RPS Negritos and 15 . 0% amongst IJV ( P = 0 . 41 ) . Whilst for hookworm infection , both prevalence and individuals with moderate to severe infections were higher in the IJV ( 26 . 2% , 41 . 0% ) versus RPS ( 18 . 7% , 24 . 0% ) ( P values = 0 . 08 , 0 . 09 ) , accordingly . The prevalence other intestinal parasitic infections ( e . g . Entamoeba sp . , Blastocystis and flukes ) was also higher in IJV versus RPS . Apart from poor hygienic behaviors as significant risk factors in both communities , low socio-economic status was highly associated with STH infections in RPS ( P<0 . 001 ) but not significantly associated in IJV . The findings showed that ex situ development plan by RPS has not profoundly contributed to the STH reduction among the OA . Conversely , burden rate of T . trichiura infections increased due to their extreme poverty and poor hygienic behaviors . Here , we are suggesting biannual mass albendazole intervention ( triple dose regimens in RPS , but a single dose in IJV ) and community empowerment to both communities . For a long-term and better uptake , these strategies must be done together with the community input and participation , respecting their traditional customs and accompanied by recruitment of more OA people in the health-care taskforce .
Cultural , socio-economic and environmental changes affect the pattern of diseases and health persistence within a population [1] . According to palaeoparasitology history , intestinal parasites , particularly soil-transmitted helminths ( STH ) had co-evolved together with humans since the very beginning , as was proven by the discovery of the parasites in an ancient fecal sample [2] . As time passed , the colonization of intestinal parasites has declined tremendously in certain parts of the world , greatly influenced by the rapid improvement of socio-economic status and sanitation standards , as seen in developed nations [3] . In contrast , however , owing to the lack of proper sanitation and good hygiene practice , intestinal parasitic infections ( IPI ) are still afflicting a majority of the deprived and underserved peoples , for example the indigenous populations in low- and middle-income countries [4 , 5] . In Malaysia , the Orang Asli ( OA ) indigenous ( which means the ‘original people’ or ‘the first people ) population has experienced a state of transition between hunter-gatherer lifestyles with strong adherence with the tropical rain forest towards modernity . The transition took place since the initiation of a redevelopment program among OA communities in 1970s by the Malaysian government , with the aim to demarginalize the OA [6 , 7] . The program was primarily conducted by improving existing OA villages ( in situ development ) and initiating resettlement plans ( Rancangan Pengumpulan Semula or RPS ) ( ex situ development ) . Under the RPS program , the OA were regrouped and relocated to new resettlements at the peripheries or near township areas with the aim to improve their health by having better access to health-care facilities , increasing their socio-economic status and education opportunities . Basic facilities such as housing , electricity and water supply , and schools were provided . Customary cultures such as foraging , hunting and subsistence cropping were gradually replaced by cash crop agriculture ( e . g . palm oil or rubber plantations ) . These RPS communities consequently developed a closer contact with other mainstream communities ( e . g . the Malay community ) [6 , 7] . In 2011 , the fraction of the OA communities that have been relocated under the RPS program was 63 . 0% [8 , 9] . On the other hand , some tribes of OA still live and remain in the inland within the forest areas ( 37 . 0% ) and still prefer the foraging and gathering lifestyles despite provision of basic infrastructure in their villages ( in situ development ) [7 , 9] . While the transition towards modernity by the rapid growth of socio-economic advancement has shown significant reduction of STH in the general population , such improvement is negligible among the OA communities [10 , 11] , since STH infections are still highly endemic [12–14] . Given these contrasting findings , the pre-assumptive idea that demarginalization and redevelopment among OA lead to a better standard of living and consequently facilitate the reduction of intestinal parasitism among this community need to be reevaluated [10 , 15] . While most of the prevalence studies highlighted low household income and sanitary behavior as the significant risk factors for STH infections among OA [12 , 16 , 17] , very few focused on the effect of demarginalization and resettlement on the infections . Therefore , our ultimate objective is to address the latter aspect , which could be achieved primarily by comparing the findings from two different OA communities from similar tribes which have different environment and ecological settings . In the present study , from three OA tribes; Senoi , Proto-Malay ( or Aboriginal-Malay ) and Negrito [18 , 19] , the Negrito was chosen as our study population based on three aspects; the tribe is the earliest inhabitants in Peninsular Malaysia ( ~50 , 000 years ago ) [20] , have the least population ( ~5000 ) currently and are known as hunter gatherers [19] . Moreover , except for a few reports available on the Negritos , the majority of the STH infection studies were conducted mainly among other tribes , namely the Senoi and Proto-Malay due to their larger population sizes [16 , 21 , 22] . Given these factors , we believed the Negritos provide a good model in determining how transition towards demarginalization influences the pattern of STH infections that traditionally inflict them [21] . Here , we report a finding pertaining on the current pattern of intestinal parasitism , STH infection intensity and associated risk factors by comparing the Negritos who are still living in the inland jungle ( in situ improvement ) and those who have undergone resettlement ( RPS ex situ development ) at the peripheries of towns .
This study received ethical approval from both the Ethics Committee of the Universiti Teknologi MARA ( reference no: 600- IRMI ( 5/1/6 ) and National Medical Research Register ( NMRR ) , Ministry of Health , Malaysia ( NMRR-17-3055-37252 ( IIR ) . Permission to conduct the study was obtained from the Department of Orang Asli Development ( JAKOA ) [Reference no: JAKOA/pp . 30 . 052J1d9 ( 29 ) ] . A cross-sectional survey with a convenient , snowball sampling method was carried out from May 2016 to April 2017 in eight villages comprising all six Negrito sub-tribes; Jahai , Bateq , Kintak , Lanoh , Kensiu and Mendriq . The villages were selected based on the list of Negrito villages permitted by JAKOA , by further invitations from headman of the tribes , and after agreement and willingness of participation by each headman and members of their village . These villages are all located in the northern states of Peninsular Malaysia ( S1A Fig ) . Prior to sample collection , visits and meetings were conducted at the selected villages . A short briefing was held before each sampling session , during which the purpose and method of the study were clearly informed to prospective participants . Those who agreed to participate either signed or thumb-printed informed consents , witnessed by the accompanying JAKOA officer ( s ) . Permission and consent via signature from legal representatives were obtained for participants below 12 years old . Participants were informed of their rights to withdraw from the study at any time without prior notice . For the purpose of comparison , the villages were grouped into two categories based on location and types of development; ‘Inland Jungle Villages ( IJV ) of in situ improvement’ ( S1B Fig ) and ‘Resettlement Plan Scheme ( Rancangan Pengumpulan Semula or RPS ) of ex situ development’ ( S1C Fig ) . The IJV community refers to the group of Negritos staying in inland tropical rain forest ( as in the case of Lanoh , Kintak and some groups of Batek and Jahai ) with some still practicing hunting and gathering lifestyles . Each village has a population size from as low as 50 to as high as 300 individuals and were not easily accessible , except by boat and four-wheel-drive vehicles . Despite being officially recognized by the government , the villages lacked basic facilities , some lacking electricity due to their very remote locations . Because contact with outsiders was minimal , their transition towards demarginalization is regarded as slow . Meanwhile , the RPS community is defined as groups of Negritos which have gathered and relocated into designated settlements at the peripheries of towns ( these Kensiu , the Mendriq and some groups of Batek and Jahai ) by JAKOA . Each resettlement has a population size of between 170 and 450 people . Their locations are near the road , easily accessible and their communities are provided with better housing , piped water supply , electricity , education and health-care facilities . Their transition towards assimilation and demarginalization with mainstream populations within the country is faster . A detailed summary of the sub-tribes and villages included in this study is shown in Table 1 . The sample size for this study was determined according to the formula by Wang and Chow [23] . The calculation was performed using PS Power and Sample Size Calculation Software for two proportions with the following parameters; 5% level of significance at 95% confidence interval , 80% power of study and anticipated STH infection prevalence as follows: 81 . 3% among RPS [16] and 96 . 5% among IJV Negritos [21] . By adjusting for 25% attrition rate , the minimum number of participants required in this study was estimated at 160 ( 80 participants from each population category ) . We managed to voluntarily recruit a total of 430 Negritos in this study . However , only 416 of these ( representing approximately 8 . 3% of the entire Negrito population in Malaysia ) which had paired stool samples and datasets were included in the analyses ( Table 1 ) . Out of the 416 participants , 149 ( 35 . 8% ) Negritos were categorized under IJV and the remaining 267 ( 64 . 2% ) belonged to RPS . Interviews were conducted by two trained research assistants in the Malay language based on a pretested questionnaire . The questions were mainly on participants’ demographic data ( i . e . sub-tribes , age , gender and number of family members ) , socio-economic and educational status ( i . e . household monthly income , occupational status and level of education ) , general sanitation and environmental conditions , which include source of water supply and availability of latrine system ( pour/flush toilet or pit latrine ) . The source of water supply was categorized into treated ( government piped water ) and untreated ( river , lakes , mountain water , etc . ) . Other questions were on behavioral risks ( personal hygiene ) such as habit of washing hands after defecation or after playing/in contact with soil , indiscriminate defecation ( preference to defecate anywhere without any specific location or latrine , usually in the bushes or near the river ) , and closer contact ( own or always playing ) with domestic animals ( commonly dogs and cats ) . Finally , we also asked about their previous history of anti-helminthic treatment; whether they have taken any anti-helminthic drug ( prescribed by medical personnel or during a deworming program ) or none at all during their entire life . For young children , the information was collected by interviewing their parents or guardians in their home setting . Those with a history of antibiotic or anti-helminthic treatment in the previous 3 months before the commencement of the study were excluded from this study . Pre-labelled capped stool containers were given to the participants a day before stool sample collection along with instruction on correct placement of their stool into the containers . Stool samples were collected without age discrimination , the following morning between 8 and 11 am . Each stool sample was divided into three aliquots [fresh , 10% formalin ( only a few drops of formalin so that the samples do not became liquefied ) , and 2 . 5% potassium dichromate] , sealed with zip-locked plastic bag , kept in cool boxes during transportation , transferred to a laboratory in the Institute for Medical Molecular Biotechnology ( IMMB ) , Universiti Teknologi MARA within 6 to 12 hours of duration and stored at 4 °C before examination . The laboratory examination was conducted using different methods as soon as possible upon the arrival of samples to the laboratory . The samples were first processed using a standard direct wet smear and formalin–ether sedimentation [25] to detect the presence of ova or cyst in the stool samples . A modified Kato–Katz method was then applied to quantify the burden of STH infections within 4 hours of reaching the laboratory . Duplicate 41 . 7 mg Kato–Katz thick smears were prepared from each fresh and formalin-fixed sample ( the preservative was discarded first ) , and were read twice after 15 minutes by two different technologists . The corresponding results were compared and in cases of significant inconsistencies ( positive vs . negative and/or difference in egg counts of more than 20% ) , the slides were re-examined . The total number of eggs was multiplied by a factor of 24 ( number of eggs × 24 ) to produce the number of eggs per gram ( epg ) of feces . The lower limits of moderate and heavy infections were 5 , 000 and 50 , 000 epg for A . lumbricoides; 1 , 000 and 10 , 000 epg for T . trichiura; and 2 , 000 and 4 , 000 epg for hookworm [26 , 27] . Modified Wheatley Trichrome ( according to manufacturer’s protocols—Thermo Scientific Remel Trichrome Stain Kit ) and modified acid-fast staining techniques ( adapted from CDC DPDx stool specimens—staining procedure ) were also incorporated in this study to facilitate detection and provide accurate identification of intestinal protozoa . Stool samples were considered positive if the intestinal parasites were detected by any of these methods . Data were double-entered , cross-checked and merged into a single data set in a Microsoft Office Excel 2010 document . Statistical analysis was performed using IBM SPSS version 20 ( SPSS , Chicago , IL , USA ) . Demographic data , socio-economic , environmental and behavioral factors were treated as categorical variables . For the number of family members , we categorized them into <7 members and >7 members as it was generally common among OA to have large family sizes . Monthly household income was categorized into ~ <USD 125 and >USD 125 because more than half of the studied populations have a gross monthly income of less than USD 125 . Descriptive statistical analyses were performed to obtain a clear understanding of the population . Frequency , percentages ( rate ) , measures of central tendency ( means , medians and other percentiles ) and dispersion ( standard deviations , ranges ) were computed to describe the characteristics of the studied population . Shapiro–Wilk statistic coupled visual diagrams ( Q–Q Plots and histogram ) were used for assessing the normality of the scores . P-value of more than 0 . 05 indicates normality . For normally distributed data , arithmetic means with standard deviations ( SD ) or standard error of the mean ( SE ) were presented . For data not normally distributed , medians with data range ( interquartile , IQR ) were used . Pearson’s chi-squared test ( χ2 ) or Fisher’s exact test were used to determine the independence between categorical independent variables and the outcome . The same test was used to measure the association of STH infections with the test variables . An independent t test was calculated by analyzing means from two independent continuous variables . For STH risk measurements , odds ratio ( OR ) and 95% confidence interval ( CI ) were analyzed by univariate logistic regression analysis . All variables with P-value less than 0 . 25 in univariate analysis were accepted for further multivariate ( backward stepwise ) logistic regression [28] . Model fitness was determined by the Hosmer–Lemeshow statistic . A P-value of less than 0 . 05 ( P < 0 . 05 ) was taken as significant .
A total of 416 Negritos ( 149 IJV and 267 RPS ) participated in this study with ages ranging from 2 to 64 years old . By age groups , the majority of the respondents ( 47 . 1% ) were primary school children aged between 7 and 12 years old . The gender ratio was 1:1 . Table 2 shows the demographic profiles , socio-economic , environmental conditions and behavior characteristics in our general studied participants and according to the categories ( IJV versus RPS ) . With regards to the IJV Negritos , even though their locations are far from the township areas , they had significantly better socio-economic status ( 40 . 9% with a household income >USD 125 ) compared with those living in the RPS community ( 29 . 6% ) . However , about 20 . 4% children were not enrolled in school with only some parents sending their young children to the provided boarding school for OA children due to distance factor . In terms of hygiene behavior , walking barefoot was common and they also preferred to defecate indiscriminately ( open defecation without any specific locations or at any designated latrines ) mainly at the river and bushes . Moreover , about 50 . 3% had previously never taken or been prescribed with any anti-helminthic treatment ( AHT ) . Despite better development in the RPS community , the majority of the houses did not have electricity supply due to their inability to settle outstanding bills . Similar situations were seen with their piped water supplies . As a consequence , rivers ( 54 . 7% ) remained the main source of water supply , especially for bathing and washing . A single latrine was commonly available in the RPS community , but the facility was not fully utilized because most of them still preferred to defecate indiscriminately as practiced in IJV . Education attainment was better in RPS , with most children being enrolled in school . However , not many of their adolescents continued or completed secondary school . Regarding drug treatment , they were highly exposed , with 74 . 9% having taken at least one dosage of AHT due to easier access to health-care facilities and periodic visits by medical personnel . Of the 416 participants , the overall prevalence of IPIs was 87 . 0% , with 358 ( 86 . 1% ) found to be infected with at least a single species of STH and 106 ( 25 . 5% ) with at least one species of intestinal protozoa ( Table 3 ) . By community categories , the prevalence of STH infections was found to be significantly higher in the IJV community compared to RPS ( P = 0 . 02 ) . The most dominant STH species was T . trichiura ( IJV: 72 . 5%; RPS: 71 . 9% ) , followed by A . lumbricoides ( IJV: 40 . 3%; RPS: 44 . 9% ) and hookworm ( IJV: 26 . 2%; RPS: 18 . 7% ) . The prevalence of polyparasitism ( having more than one type of infection ) was slightly higher in the IJV community than in the RPS community but not significant . The percentage of participants positive with Entamoeba sp . , Blastocystis sp . and other types of parasites ( flukes , tapeworms , unidentified eggs and cysts ) was greater in IJV ( P = 0 . 01 ) indicating higher diversity of parasitic infections among the IJV community . Our findings showed that Negritos who live in IJV had 2 . 1 times greater risks ( 95% CI: 1 . 1 , 4 . 1; P = 0 . 02 ) to be infected with STH than RPS . We then compared the associations between STH infections with demographic , socio-economic and environmental circumstances , and personal hygiene behavior variables ( Table 5 ) . In the IJV community , univariate analysis indicated significant association between children aged <12 years old ( P = 0 . 03 ) , persons who defecated indiscriminately ( commonly at the bushes ) ( P = 0 . 04 ) , walking barefoot ( P = 0 . 003 ) and had close contact with domestic animals ( P = 0 . 002 ) with the presence of STH . Meanwhile in RPS , the Negritos with low household monthly income ( <USD 125 ) had 3 . 9 times greater risks ( P <0 . 001 ) to be infected with STH . Other significant risk factors in RPS communities were related to poor sewage disposal system ( P = 0 . 01 ) and bad hygiene behaviors , such as not washing hands after playing or contact with soil ( P<0 . 001 ) , not washing fruits and vegetables before eating ( P = 0 . 03 ) and closer contact with animals ( P< 0 . 001 ) . Table 6 shows the result of further multivariate models for final significant predictors for STH infections in both the IJV and RPS communities . For the IJV community , only three factors; closer contact with animals , indiscriminate defecation , and walking barefooted ( P values: 0 . 01 , 0 . 03 and 0 . 04 , accordingly ) which were all grouped under poor hygiene behaviors , remained as the significant predictors . Goodness-of-fit by the Hosmer–Lemeshow test indicated the model fits the data well ( χ2 = 2 . 20; df = 4 , P = 0 . 69 ) . With regards to RPS community , from eleven variables , five factors were retained; low monthly income of less than USD 125 ( P < 0 . 001 ) , improper sewage disposal ( P = 0 . 03 ) , poor hygiene behavior of not washing hands after playing with soil ( P = 0 . 01 ) , indiscriminate defection ( 0 . 04 ) and those with a closer contact with animals ( or have domestic animals in their household ) ( P < 0 . 001 ) had a higher prevalence rate of STH than their counterparts with STH infections ( Hosmer–Lemeshow test: χ2 = 5 . 83 , df = 7; P = 0 . 56 ) .
There are a few limitations that need to be considered while interpreting the present findings . Firstly , the findings were based on one time point collection within a certain period without being able to identify the relationship of current infection with previous and future infections . Secondly , the study population was quite small and represented only 8 . 3% of the total Negrito population . Therefore , it may not reflect the community of OA as a whole . Thirdly , the sampling was based on convenient , snowball method and this may provide some biased outcomes , as the sampling is not random . Nevertheless , the sampling approach was the most appropriate and feasible considering the challenges and strict procedures to recruit the OA participants , especially in the IJV communities . In addition , the prevalence of S . stercoralis could be underestimated because all the methods used have a low sensitivity for this particular parasite . Despite a slight reduction of intestinal parasitism in the RPS , the ex situ development plan has not profoundly contributed to a positive impact on the status of helminth infestation among the OA . In fact , the burden of T . trichiura was more intense due to further poverty , adoption of similar poor hygienic behaviors and lack of proper sanitation . Nevertheless , the concept of RPS for demarginalization and consequently reduction of STH endemicity , can still be further improved taking into consideration holistic development of proper sanitation and socio-economic improvement , biannual mass deworming programs , and more importantly coupled with health educations strategies to change mindsets in both RPS and IJV communities . For long-term intervention and better uptake , these strategies must inclusive , with the participation and empowerment of the OA of the respective community , respecting their traditional customs and accompanied by recruitment of more OA people in the health-care taskforce . Importantly , the increased involvement of OA people in policy-making and political affairs which are related to the changes of ecological , social and economic drivers among them may indirectly improve their health and consequently reduce the burden of helminth infections . Otherwise , the malady among them will remain unchanged and unsolved , increasing the cost of economic burden in controlling these infections . | Pattern of diseases in a community are associated with changes of culture , socio-economic and environmental conditions . In Malaysia , while transition towards modernity by socio-economic development has shown significant reduction in intestinal parasitism in the general population , no significant difference was observed among the Orang Asli ( indigenous ) communities . Nevertheless , since the 1970s , the Malaysian government has initiated the Resettlement Plan Scheme ( RPS ) to improve health , socio-economic status and education opportunities of the OA . Known as hunter gatherers , the Negritos represent a good model to study the effects of living conditions transition on intestinal parasitism . Like other tribes , they are also undergoing the transition from foraging and gathering practices towards demarginalization primarily by resettlement . Thus , the present study aims to evaluate STH infections in two categories of Negritos; ( i ) Inland Jungle Villages ( IJV ) communities , living in the interior forested areas and ( ii ) the Negritos who have been gathered and relocated to designated settlements closer to towns ( RPS ) . A study to compare the current pattern of STH profiles in terms of prevalence and intensity of infection together with their associated risk factors among these two categories of communities was conducted . Since STHs are still plaguing the OA communities , the information gathering in the present study will be crucial for customizing a cost-effective deworming program involving selection of drug dosages , frequency of administration , type of modalities and most importantly community inclusion and empowerment programs through health education . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"water",
"resources",
"helminths",
"tropical",
"diseases",
"hookworms",
"parasitic",
"diseases",
"animals",
"physiological",
"processes",
"ascaris",
"ascaris",
"lumbricoides",
"neglected",
"tropical",
"diseases",
"medical",
"risk",
"factors",
"natural",
"resources",
"epidemiology",
"defecation",
"helminth",
"infections",
"eukaryota",
"physiology",
"nematoda",
"biology",
"and",
"life",
"sciences",
"soil-transmitted",
"helminthiases",
"organisms"
] | 2019 | Prevalence, intensity and associated risk factors of soil transmitted helminth infections: A comparison between Negritos (indigenous) in inland jungle and those in resettlement at town peripheries |
Begomoviruses are exclusively transmitted by whiteflies in a persistent circulative manner and cause considerable economic losses to crop production worldwide . Previous studies have shown that begomoviruses accumulate in vesicle-like structures in whitefly midgut cells and that clathrin-mediated endocytosis is responsible for their internalization . However , the process by which begomoviruses are trafficked within whitefly midgut cells remains largely unknown . In this study , we investigated the roles of vesicle trafficking in the transport of Tomato yellow leaf curl virus ( TYLCV ) , a begomovirus that has spread to over 50 countries and caused extensive damage to a range of important crops , within midgut cells of whitefly ( Bemisia tabaci ) . By disrupting vesicle trafficking using RNA silencing and inhibitors , we demonstrated that the early steps of endosomal trafficking are important for the intracellular transport of TYLCV in the whitefly midgut . In addition , our data show that , unlike many animal viruses , TYCLV is trafficked within cells in a manner independent of recycling endosomes , late endosomes , lysosomes , the Golgi apparatus and the endoplasmic reticulum . Instead , our results suggest that TYLCV might be transported directly from early endosomes to the basal plasma membrane and released into the hemolymph . Silencing of the sorting nexin Snx12 , which may be involved in membrane tubulation , resulted in fewer viral particles in hemolymph; this suggests that the tubular endosomal network may be involved in the transport of TYLCV . Our results also support a role for the endo-lysosomal system in viral degradation . We further showed that the functions of vector early endosomes and sorting nexin Snx12 are conserved in the transmission of several other begomoviruses . Overall , our data indicate the importance of early endosomes and the tubular endosomal network in begomovirus transmission .
Insects transmit the majority of plant viruses [1–3] . The process of virus transmission by an insect vector varies based on how the virus is acquired , retained , and inoculated into plants . Some plant viruses , such as members of the genera Caulimovirus , Cucumovirus and Potyvirus , are transmitted in a non-circulative manner , in which the viruses are acquired by feeding on infected plants , retained on the surface of the stylet , food canal , or foregut , and then inoculated into new host plants [4–6] . The coat proteins or non-structural proteins of some of these viruses aid their retention on the inner cuticular lining of the vector feeding apparatus via interactions with specific receptors [7–9] . By contrast , persistent circulative viruses have developed much more complex interactions with their vectors . These viruses need to travel from the gut lumen into the hemolymph , move to the salivary gland , and finally be secreted from the salivary gland into new host plants [10 , 11] . Barriers to this transmission exist in the midgut and salivary gland , in which septate junctions are formed among epithelial cells to control the exchange of substances between hemolymph and midgut or salivary gland [12] . In many cases , circulative viruses can be transmitted at higher rates after injection into the vector hemocoel than after oral acquisition , suggesting that delivery across the insect gut is a considerable barrier to virus transmission [13 , 14] . Understanding the cellular processes that viruses exploit to facilitate their transport within vectors remains challenging . Animal viruses enter cells either via direct penetration through the plasma membrane or via endocytosis . The majority of animal viruses enter cells by hijacking the host cell’s clathrin-mediated endocytosis pathway [15] . Viral particles are then transported to early endosomes and sorted to various intracellular destinations [15 , 16] . After arrival in the lumen of specific vesicles , environmental cues such as acidified pH , changes in the redox environment , and proteolytic cleavage induce conformational changes in viral proteins and activate viral penetration into cytoplasm [17–19] . For instance , a significant proportion of Adeno-associated virus ( AAV ) particles are transported to the trans-Golgi network ( TGN ) , whereas Ebola virus particles need to be transported from endosomes to lysosomes to complete their infection cycle [20 , 21] . However , the intracellular routes plant viruses take within their vectors may differ significantly from those animal viruses take . How insect-vectored plant viruses exploit vector cells for delivery across the midgut is poorly understood . In the past three decades , begomoviruses ( genus Begomovirus , family Geminiviridae ) , one of the most important groups of plant viruses in tropical and subtropical regions , have caused considerable economic losses to a variety of crops [3 , 22–24] . Begomoviruses are transmitted exclusively by the whitefly Bemisia tabaci in a persistent circulative manner [3] , and the emergence of begomoviruses as important pathogens is closely related to the increasing prevalence of whiteflies worldwide [25 , 26] . A previous immuno-electron microscopy study using gold-labeled secondary antibodies showed that the begomovirus Tomato yellow leaf curl virus ( TYLCV ) can be detected in vesicle-like structures in the midgut cells of whitefly [27] . Pan et al . [28] further showed that TYLCV can be internalized into midgut cells through the clathrin-dependent pathway and that the endosomal system may play an important role in virus transport across the whitefly midgut . These results suggest that membranous vesicles are recruited by begomoviruses for their entry into vector midgut cells . However , the process by which these viruses are trafficked in epithelial cells remains largely unknown . Many questions , including how begomoviruses are transported to the basal membrane of the epithelial cell and secreted into hemolymph , have not been answered . In this study , we explored begomovirus trafficking in whitefly midgut cells using immunofluorescence , inhibitors , and double strand RNA ( dsRNA ) -mediated RNA silencing . We showed that only early steps of endosomal trafficking are involved in the intracellular transport of begomoviruses , suggesting that viral particles might be transported directly from early endosomes to basal membranes .
Animal viruses make extensive use of vesicle trafficking following endocytosis in their infection processes [15 , 17] . The Arp2/3 complex can facilitate endocytosis and vesicle trafficking through its functions in organizing actin filaments and regulating polymerization [29–31] . To confirm their roles in the transmission of TYLCV by whitefly , Arp2 and Arp3 were subjected to dsRNA-mediated gene silencing analysis . The efficiency of dsRNA-mediated gene silencing was confirmed at the mRNA level by quantitative RT-PCR ( Fig 1A ) . Silencing of Arp2 or Arp3 resulted in significant decreases in the quantity of TYLCV acquired by whiteflies ( Fig 1B ) . Next , whether Arp2 and Arp3 influenced the transport of viral particles across the midgut barrier was examined by quantifying virus abundance in the hemolymph . Virus abundance in hemolymph was significantly lower in the treatment groups than in the control groups ( Fig 1C ) . To rule out the possibility that silencing Arp2 and Arp3 influences the ingestion of virus-containing phloem sap by whiteflies , the honeydew of whiteflies was collected and its total sugar content was quantified by the anthrone reaction . No significant difference was observed between the control and treatment groups ( Fig 1D ) , suggesting that silencing Arp2 or Arp3 did not affect virus ingestion . To further investigate the role of vesicle trafficking in TYLCV transmission , we fed whiteflies with Brefeldin A ( BrefA ) , which can disrupt intracellular vesicle transport by a pleiotropic effect on the entire endosome system and the Golgi apparatus [32] . BrefA treatment induced the mislocalization of the trans-Golgi network ( S1 Fig ) , confirming its effects in the whiteflies , and BrefA treatment did not affect whitefly feeding ( Fig 1E ) . The acquisition of TYLCV by whitefly was significantly inhibited after BrefA treatment ( Fig 1F ) . Feeding of BrefA inhibited the transport of virus across the midgut and led to a significant reduction in hemolymph virus abundance after 24 and 48 hours acquisition access period ( AAP ) on an infected plant ( Fig 1G ) . Moreover , when whiteflies were used to transmit TYLCV from infected to uninfected host plants , BrefA treatment significantly decreased their transmission efficiency ( Fig 1H ) . Overall , our data suggest that vesicle trafficking is critical for the transport of TYLCV into vector hemolymph . The structure of the whitefly alimentary canal and filter chamber are shown in Fig 2A , and the two arms that join the filter chamber are gastric caeca [33] . Fig 2B and 2C show cross sections of the midgut . The whitefly midgut has only one layer of epithelial cells and is rich in microvilli . In thin sections of whiteflies exposed to the virus for a 7d AAP , aggregates of TYLCV-like particles , ca . 19 nm in diameter , were consistently found in epithelial cells and were bound by a single membrane ( Fig 2D–2I ) . These accumulations were mostly found in paracrystalline arrays , which is similar to the form Maize streak virus ( MSV ) takes in its leafhopper vector [34] . The vesicles containing these aggregates were not limited to any particular region of epithelial cells , but could be found in the apical region close to microvilli ( Fig 2D and 2E ) , close to the basal plasma membrane ( Fig 2F and 2G ) , or between these locations ( Fig 2H and 2I ) . In control whiteflies , TYLCV-like particle aggregates were never found in midgut cells ( Fig 2J and 2K ) . However , whether these aggregates consisted of TYLCV particles remains uncertain . Subsequently , we used immunofluorescence to localize the virus in the midgut , in relation to lectins used as organelle markers . Lectins are oligomeric proteins with saccharide-binding sites that can recognize and bind particular sugar molecules . Usually , specific oligosaccharides are associated with a certain organelle , and lectins can thus serve to identify cellular components . Lectin WGA binds to N-acetylglucosaminyl residues and is used for staining the nuclear core , plasma membrane and sarcolemma [35–37] . Lectin GS-II is used for staining intermediate-to-trans Golgi because of its affinity for α- and β-N-acetyl-D-glucosaminyl residues [38] . Lectin HPA selectively binds to type A erythrocytes and to α-N-acetylgalactosaminyl residues found in the cis-Golgi [39] . The midgut of whiteflies that had a 72 h AAP on TYLCV-infected plants were dissected and prepared for immunofluorescence . Confocal microscopy showed no appreciable colocalization between TYLCV and WGA or GS-II ( Fig 3A ) . However , HPA lectin specifically labelled spherical structures within midgut cells , and almost all TYLCV colocalized with the HPA-labeled structures . In the enlarged image , HPA lectin labels a vesicle-like structure surrounding TYLCV ( Fig 3A ) . Previous studies showed that BrefA can impair the functions of various vesicles [20 , 32] . We next investigated the effect of BrefA on the HPA-labeled structures and TYLCV . Whiteflies that had been fed with BrefA were transferred to TYLCV-infected plants for 72 h and prepared for immunofluorescence . BrefA had no visible effect on the formation of HPA-labeled structures; however , it disrupted the colocalization between HPA-labeled structures and TYLCV ( Fig 3B ) . This indicates that BrefA treatment can lead to disrupted localization of TYLCV . Next , whiteflies were prepared for immunofluorescence after different AAP on TYLCV-infected plants . The accumulation of virus in HPA-labeled structures was observed as early as 1 day after transfer to infected plants . The quantity of virus in midgut cells gradually increased over time , and most viruses were localized within HPA-labeled vesicles from day 1 to day 8 ( Fig 4 ) . The strict colocalization between viral particles and HPA-labeled vesicles indicates that the majority of TYLCV accumulated in a single kind of vesicle and few ( if any ) particles were released into the cytoplasm . Since it has been reported that the terminal α-N-acetylgalactosaminyl residues that lectin HPA selectively binds to are added in the cis-Golgi and then substituted in the trans-Golgi [39] , our results suggest that , after internalization , TYLCV might localize in the Golgi apparatus . Retromer complex plays important roles in retrograde transport of cargoes to the Golgi apparatus , and its subunits Vps26 , Vps29 and Vps35 are responsible for cargo selection [40] . To verify the role of the Golgi apparatus in TYLCV transport , these genes were selected as targets for RNA interference , and we also used Golgicide A ( GolA ) , a highly specific inhibitor of the Golgi apparatus . Quantitative RT-PCR showed that Vps26 , Vps29 and Vps35 were successfully silenced ( S2 Fig ) . The effect of GolA on the Golgi apparatus was confirmed by the dispersal of the trans-Golgi network ( S2 Fig ) . Measurement of sugar content in honeydew indicated that silencing of these genes or GolA treatment did not influence whitefly feeding ( S2 Fig ) . However , to our surprise , neither GolA treatment nor silencing of retromer complex genes inhibited virus acquisition by whitefly ( Fig 5A and 5B ) , suggesting that the Golgi apparatus may not be involved in the transport of virus . This result is surprising because retromer complex is essential for retrograde transport to the Golgi , and TYLCV was found to be colocalized strictly with an HPA-labeled structure , which is predicted to be the cis-Golgi apparatus [40 , 41] . To address this apparent discrepancy , midguts of virus-carrying whiteflies were further labeled using antibodies against marker proteins for the Golgi apparatus and endoplasmic reticulum ( ER ) [42] . Surprisingly , no appreciable colocalization was detected between virus and the Golgi apparatus or ER ( Fig 5C ) . We further investigated the spatial relationship between the Golgi apparatus marker protein and HPA-labeled structures and found no appreciable colocalization ( Fig 5D ) . Taken together , our results suggest that TYLCV virus particles accumulated in some kind of vesicle labelled by HPA , but not in the Golgi apparatus or ER . To investigate what kind of vesicle this is , we conducted lectin affinity capture followed by mass spectrometry to identify proteins on virus-containing vesicles labeled by lectin HPA . Five proteins were captured by lectin HPA: glycosylphosphatidylinositol ( GPI ) -anchored glycoprotein , myosin heavy chain , beta-actin , lysosomal associated membrane protein-1 ( Lamp1 ) and sarcoplasmic/endoplasmic reticulum calcium-transporting ATPase ( SERCA ) . GPI-anchored glycoprotein is localized mainly on the plasma membrane but can be endocytosed via caveolae or other plasma membrane components and then transported to endosomes [43] . The cytoskeleton and motor proteins are critical to vesicular movement , localization , and membrane fission . We speculated that myosin heavy chain and beta-actin were detected due to their interactions with proteins on vesicles , not their direct glycosylation [44 , 45] . Lamp1 is a glycoprotein that is primarily associated with the endosome and the lysosome , and SERCA may be responsible for calcium entry into lysosomes [46 , 47] . This suggests that viral particles may accumulate in endosomes or lysosomes following their internalization . To determine the roles of endosomes in virus trafficking , we first examined whether TYLCV localizes within endosomes using specific antibodies against marker proteins Rab5 ( early endosome ) , Rab7 ( late endosome ) and Rab11 ( recycling endosome ) . We found that some TYLCV particles were colocalized with Rab5-labeled early endosomes ( Fig 6A ) . However , we found no appreciable colocalization between TYLCV and Rab7 or Rab11 ( Fig 6B and 6C ) . Knock down of Rab5 , Vps8 ( a specific subunit of the endosomal CORVET complex ) , Vps11 and Vps33a ( two subunits shared by CORVET and the HOPS complex ) suppressed the acquisition of virus and inhibited the delivery of TYLCV across the midgut ( Fig 7 ) . Whereas silencing Mon1 , Rab7 , Rab11 , Lamp1 , and Vps39 ( a specific subunit of the HOPS complex ) had no influence on the virus titer in whitefly ( Fig 7 ) . For each gene , silencing was confirmed by quantitative RT-PCR and was shown to have no detectable influence on phloem sap ingestion by quantifying sugar content in honeydew ( S3 Fig ) . The small GTPase Rab5 can interact with CORVET tethering complexes and regulate the early endocytic pathway by mediating the fusion of early endosomes with endocytic and Golgi-derived vesicles [48 , 49] . Rab7 , Mon1 and the HOPS complex are responsible for endosomal maturation and membrane fusion at late endosomes and lysosomes [45 , 50 , 51] . The small GTPase Rab11 controls the traffic of recycling endosomes , which are responsible for delivering cargoes to the plasma membrane in a slow recycling route [52] . These results indicate that the early steps of endosomal trafficking play important roles in virus transport . The maturation of early endosomes into late endosomes is coupled with acidification , which is essential for hydrolytic activity , membrane trafficking , and cargo sorting [53–55] . The exposure of viral particles to low pH or hydrolytic activity in endo-lysosomal system can trigger conformational changes in the structural proteins of viruses , which is a necessary step in some viral life cycles [18 , 56] . In order to study whether endosomal maturation and acidification participate in the trafficking process of TYLCV , we used chloroquine ( Chloq ) to inhibit acid flux and prevent the acidification of endo/lysosome . Rather than inhibiting viral transport , feeding whiteflies with Chloq resulted in a significant increase in virus abundance in two of three independent experiments ( Fig 8 ) . This result indicates that instead of aiding the transmission of TYLCV , the maturation and acidification of the endo-lysosomal system caused the proteolytic degradation of viral particles . To further investigate the architecture of the virus-containing vesicles , we used a single-molecule imaging method , stochastic optical reconstruction microscopy ( STORM ) , to visualize the ultrastructure of these vesicles and their relationship with TYLCV . In STORM images , the empty circular or oval-shaped structures labeled by HPA lectin are speculated to be the membranes of vesicles . The STORM images clearly showed that TYLCV is encapsulated by these HPA-labeled structures ( Fig 9A ) . Three-dimensional STORM images further showed that the vesicles form an ellipsoid-like structure surrounding viral particles ( Fig 9B ) . A full 3D visualization of the HPA-labeled vesicle is provided in S1 Movie . We subsequently studied the morphologic variability of the HPA-labeled vesicles . In total , three kinds of vesicles were found: large vesicles with diameters ranging from 0 . 5 to 2 μm , which show great pleomorphism ( Fig 9C–9E ) ; small vesicles with diameters of 200 nm and precisely spherical shapes ( Fig 9E and 9F ) ; and vesicles with a tubular shape , which might be part of the tubular endosomal network ( Fig 9G and 9H ) . Viral particles were associated with all three types of vesicles as well as with the basal plasma membranes of midgut cells ( Fig 9C ) . These results further confirmed that TYLCV virus particles are transported within vesicles in the cells of the whitefly midgut . Interestingly , most viral particles were localized close to the perimeters of vesicles ( Fig 9A–9E ) , indicating that viruses were recruited to vesicle membranes , likely by receptor-ligand interactions , rather than suspended in the fluid within vesicles . The tubular vesicles may be responsible for sorting virus to the plasma membrane or other destinations [57 , 58] . The cargoes in endosomes are sorted depending on tubular-shaped membrane compartments , which allow the packaging of mainly membrane-bound components separately from luminal contents [58] . These tubules are induced and stabilized through a phosphatidylinositol-3-monophosphate ( PtdIns3P ) binding protein family , Snx [59] . We searched the whitefly genome and identified four Snx genes: Snx2 , Snx4 , Snx6 and Snx12 . Snx2 and Snx6 , which form a membrane-deforming subcomplex of retromer complex , are involved in retromer-mediated retrieval transport from endosomes to the Golgi apparatus; Snx4 can regulate recycling transport of the transferrin receptor to the plasma membrane; and Snx12 is homologous to Snx3 , which may be involved in the regulation of endocytosis , endosomal sorting , and signaling [59–61] . These genes were subjected to RNA silencing analysis to test their influence on virus transport . The silencing efficiency was confirmed , and the silencing had no appreciable influence on phloem sap ingestion ( S5 Fig ) . Only the Snx12-silenced whiteflies had significantly lower TYLCV viral abundance than control whiteflies after feeding on infected plants , suggesting that Snx12 could be involved in virus transport in whitefly midgut cells ( Fig 10A ) . We further examined the role of Snx12 by quantifying virus abundance in the midgut and hemolymph of whiteflies . Silencing of Snx12 caused a non-significant decrease in the quantity of virus in the midgut but significantly reduced virus abundance in hemolymph ( Fig 10B ) , indicating that Snx12 may play a role in the secretion of virus into hemolymph . The ability of whiteflies to transmit virus to new plants was also tested post-Snx12 silencing . Notably , silencing of Snx12 depressed the transmission of TYLCV by whiteflies ( Fig 10C ) . Immunofluorescence analysis further showed that virus-containing vesicles were localized to the perinuclear region after silencing of Snx12 ( Fig 10D ) . These results further validat the importance of endosomes in TYLCV transport . Finally , we tested the role of endosomes in the transmission of two additional begomoviruses closely related to TYLCV , Papaya leaf curl China virus ( PalCuCNV ) and Tomato yellow leaf curl China virus ( TYLCCNV ) [62 , 63] . Whiteflies were fed with inhibitor or injected with dsRNA as described in previous sections . Feeding whiteflies with BrefA significantly inhibited the acquisition of these viruses by whiteflies in five of six independent experiments ( Fig 11A and 11C ) . Silencing of Rab5 , Vps33a and Snx12 , which are responsible for early steps of endosomal trafficking , significantly reduced the abundance of both viruses in whitefly ( Fig 11B and 11D ) . We further observed the distribution of these two begomoviruses in the midgut of whitefly by immunofluorescence microscopy . Both viruses were colocalized nicely with HPA-labeled vesicles ( Fig 11E ) . These results suggest that early endosomes are a common route for the transport of begomoviruses in their insect vectors .
The transport of virus from the gut lumen into the hemolymph of the insect vector is an important step in the circulative transmission of plant viruses [13 , 14] . Brault and colleagues proposed that plant luteoviruses , which undergo circulative transmission , cross the aphid gut epithelium through a transcytosis process dependent on clathrin-mediated endocytosis [64] . However , the detailed trafficking processes for plant viruses in their insect vectors are poorly understood . Our experiments in the present study demonstrated that vesicle trafficking plays important roles in TYLCV intracellular transport . We showed that TYLCV accumulates in HPA-labeled structures . Silencing the CORVET complex and Rab5 can inhibit virus acquisition , suggesting that the early steps of endosomal trafficking play important roles in virus transport . In addition , the transport of TYLCV is probably independent of the Golgi apparatus , late endosomes and recycling endosomes . These results suggest that TYLCV might be transported directly from early endosomes to the basal plasma membrane in a fast recycling route . Nevertheless , the present data cannot rule out the participation of other non-classical pathways . For example , Nonnenmacher et al . [20] showed that AAV can be transported to the Golgi apparatus through a non-canonical retrograde transport pathway that is independent of the retromer complex . Thus , TYLCV may be transported to the Golgi apparatus through an unknown pathway and then rapidly secreted to the hemolymph once it arrives . Many viruses need to be delivered between several kinds of vesicles during their infection process . However , we showed that the transmission of TYLCV was independent of the Golgi apparatus , late endosomes , lysosomes , and recycling endosomes . This suggests that the trafficking of TYLCV in vector cells might be a simpler process than the infection processes characteristic of pathogenic viruses . Pathogenic viruses need to take advantage of specific proteins or environmental conditions in vesicles to become infectious and release their hereditary material into the cytoplasm of host cells [17–19] . For example , Ebola virus particles need to be transported from endosomes to lysosomes before their cytoplasmic release [21] . After internalization , SV40 accumulates in the smooth ER as part of its productive infectious route [65] . The lysosomal cholesterol transporter protein Niemann–Pick C1 and the misfolded protein response machinery in the ER are thought to aid the escapes of Ebola virus and SV40 , respectively , from their vesicular compartments [21 , 66] . However , TYLCV and other vectored plant viruses that undergo circulative transmission have a different destination: transport into the hemolymph , followed by secretion from the salivary gland into host plants . Therefore , the delivery of virus from early endosomes to other vesicles might be unnecessary . Ultrastructural analysis of the virus-containing vesicles showed three types of vesicles that differ in the size and shape . The large vesicles may represent endosomes in which TYLCV has accumulated , and the small vesicles with spherical or tubular shapes might represent a previously described group of vesicles that bud from a donor compartment and are responsible for shuttling transport [57] . Since the basal plasma membrane but not the apical membrane of epithelial cells was labeled by lectin HPA , we hypothesize that shuttling vesicles bud from TYLCV-containing vesicles and are targeted to the basal plasma membrane or to other HPA-labeled vesicles . Interestingly , most viral particles were found near or connected to the membrane rather than being suspended in the lumen of vesicles . This indicates that TYLCV might be recruited to the membrane by specific receptor-ligand interactions and that this interaction could be important in its intracellular transport . It will be fascinating to clarify whether the same receptor is responsible for internalization and for intracellular transport . If so , this will raise the question of why and how this receptor is transported from the apical to the basal membrane of epithelial cells . If different receptors are involved , what mechanism promotes the disassociation of viral particles from the first receptor and its association with the second ? Our results also showed that inhibiting endosomal acidification leads to higher TYLCV content in whiteflies , suggesting that the virus may be transported to and degraded in lysosomes . However , no TYLCV particles were observed within late endosomes , and disrupting late endosomes or lysosomes by RNA silencing had no influence on virus content . In recent years , many studies have demonstrated that acidic degradation of extracellular cargo is not limited to lysosomes but can occur further upstream in the endo-lysosomal system [67] , suggesting that TYLCV could be degraded in early endosomes . However , because we did not have access to antibodies against lysosomes , it is still unclear which of these possibilities is correct . Taken together , our results indicate that early steps of endosomal trafficking play important roles in begomovirus transport . Based on the present experiments and the studies of others [27 , 28 , 64 , 68 , 69] , we conclude that begomoviruses are first delivered to early endosomes after clathrin-mediated endocytosis , then transported directly to the basal membrane of midgut epithelial cells . The fusion between virus-containing endocytic vesicles and early endosomes is mediated by Rab5 and the CORVET complex . Then , some of the viral particles may be transported to the basal membrane by tubular vesicles induced by Snx12 ( Fig 12 ) . Whether some viral particles are transported to lysosomes remains uncertain . To our knowledge , this is the first study regarding the intracellular trafficking of begomoviruses within their insect vectors . The exact transport pathway of begomoviruses in epithelial cells may be more complicated than the proposed model because these cells are polarized , with apical and basal domains [53] . For example , sorting to the apical and basal membranes are governed by different mechanisms and carried out by different vesicles [70] . Intriguingly , this virus is endocytosed on the apical membrane and exocytosed on the basal membrane . Observing virus transport in live cells may help future researchers better understand the process .
Middle East-Asia Minor 1 whiteflies ( MEAM1 , previously referred as the ‘B biotype’ ) were collected from Zhejiang province , China and maintained in the laboratory . Whiteflies were reared on cotton ( Gossypium hirsutum cv . Zhe-Mian 1793 ) plants in climate chambers at 26 ± 1°C with a photoperiod of 14 h/10 h and 70 ± 10% relative humidity . Every three months , the purity of the cultures was monitored by PCR-restriction fragment-length polymorphism analysis and further confirmed by sequencing of cytochrome oxidase I ( mtCOI , GenBank accession no . KM821540 ) as previously reported [71] . Infectious clones of TYLCV , TYLCCNV and PaLCuCNV were kindly provided by Professor Xue-Ping Zhou from the Institute of Biotechnology , Zhejiang University . Briefly , 1 . 4 copies of the genome of each begomovirus was cloned into a plant transformation vector and transfected into Agrobacterium tumefaciens . Tomato ( Solanum lycopersicum cv . Hezuo 903 ) plants with 3–4 true leaves were inoculated with an infectious clone of TYLCV , TYLCCNV , or PaLCuCNV to obtain virus-infected plants . They were then cultivated to the 6–7 true leaf stage for further experiments . Brefeldin A ( BrefA , Enzo Life Sciences ) was dissolved in ethanol ( Sigma ) to make a 5 mM stock solution . Chloroquine phosphate ( Chloq , Sigma ) was dissolved in sterile water to make a 700 mM stock solution . A stock solution of 50 mM Golgicide A ( GolA , Selleck ) was prepared in dimethyl sulfoxide ( DMSO , Sigma ) . All stock solutions were diluted 1:1000 in water with 30% sucrose for feeding of whiteflies . For chemical treatment , newly emerged whiteflies were collected and fed through a parafilm membrane for 24 h . In different experiments , 30% sucrose solutions containing 0 . 1% ethanol , DMSO or water were included as controls . DNA templates were generated by PCR using primers that contained the T7 promoter at both ends . Then , DNA templates and the MEGAscript T7 Transcription Kit ( Ambion , USA ) were used to synthesize dsRNA according to the manufacturer’s instructions . DsRNA was subsequently purified using phenol:chloroform extraction and isopropanol precipitation , then resuspended in nuclease-free water . The size and quality of the dsRNA were confirmed by 1% agarose gel electrophoresis , and its quantity was measured using Nanodrop ( Thermo Scientific , USA ) . Primers used for DNA template synthesis are listed in S1 Table . For RNA silencing , approximately 6 nl of purified dsRNA ( 8 μg/μl ) was injected into the thorax of each adult female whitefly using capillary and FemtoJet ( Eppendorf , Germany ) . For each gene , 200 female whiteflies were injected and dsRNA corresponding to green fluorescent protein ( GFP ) was included as control . After injection , whiteflies were kept on cotton plants for three days for recovery . Efficiency of dsRNA mediated gene silencing was verified by qRT-PCR . After inhibitor feeding or dsRNA injection , about 50 female whiteflies of each treatment group were caged on one leaf , while about 50 control whiteflies were caged on a symmetrical leave of the same virus-infected tomato plant . After 48 h AAP , these whiteflies were collected and a group of 10 female whiteflies was prepared as one sample . For each experiment , 4–6 samples were usually prepared for analysis , and the results are shown as one bar in the graphs ( n in each figure represents the number of these samples ) . The collected whiteflies were homogenized in 200 μl lysis buffer ( 10 mM Tris-HCl pH 8 . 4 , 50 mM KCl , 0 . 45% Tween-20 , 0 . 45% Nonidet P-40 , 0 . 2% gelatin and 60 mg/l proteinase K ) then incubated at 56°C for 1 h and 100°C for 10 min . The supernatant was used in quantitative PCR ( qPCR ) after centrifuging at 12 , 000 rpm for 3 min . For each inhibitor or gene silencing treatment , the experiment was repeated three times . Quantification of virus titer in midguts and hemolymph were similar . The midguts of female viruliferous whiteflies were dissected in DPBS and washed several times before use . A group of 15 midguts was prepared as one sample . For each experiment , 4 samples were usually prepared for analysis ( n in each figure represents the number of samples ) and the experiment was repeated three times . Midguts were homogenized in 30 μl lysis buffer and processed in the same way as the whitefly whole bodies . The hemolymph of whitefly was collected from the abdomen of female whiteflies using a capillary with a fine point of ~1 μm in diameter . Hemolymph of each whitefly was homogenized in 10 μl lysis buffer and processed in the same way as the whitefly whole bodies . Hemolymph from two whiteflies was pooled and prepared as one sample . For each experiment , 10–20 samples were usually prepared ( n in each figure represents the number of samples ) , and each experiment was repeated three times . Quantitative real-time PCR was used to analyze the quantity of virus acquired by the whiteflies and the efficiency of RNAi . Total RNA of whitefly was isolated using TRIzol ( Ambion , USA ) and reverse transcribed using PrimeScript RT reagent Kit ( TaKaRa , Japan ) following the manufacturer’s protocol . Quantitative PCR was performed on CFX Connect Real-Time PCR System ( Bio-Rad , USA ) using the FastStart Essential DNA Green Master ( Roche , Switzerland ) and custom-designed specific primers to the genes . Actin was used as an internal reference , and relative abundance of TYLCV or transcripts was calculated by 2-ΔCt . Primers used for real-time PCR are listed in S1 Table . After 48 h feeding on virus-infected tomato plant , the whitefly honeydew was collected by washing the leaves and inner surfaces of the micro-cages with sterile water several times . The collected honeydew was diluted to 1 ml . Glucose dissolved in water was used to produce a standard curve . The anthrone reaction was used to determine the content of sugar in the honeydew solution , as described previously [72] . Briefly , anthrone reagent was prepared by dissolving 0 . 2 g of anthrone in 100 ml 80% ( w/w ) H2SO4 . Then , 40 μl of the honeydew solution was mixed with 160 μl anthrone reagent and heated as required in a 100°C water bath . The absorbance at 620 nm was determined using a Varioskan Multimode Microplate Reader ( Thermo Scientific , USA ) . After being fed with inhibitor or injected with dsRNA , the treatment group and its control group were caged on two symmetrical leaves of the same virus-infected tomato plant for 12 h . The whiteflies were then collected and sexed . Female whiteflies were placed in groups of 4 on a true leaf of an uninfected tomato plant at the 3–4 true-leaf stage for 48 h using a micro-cage . Then , whiteflies were removed from tomato plants , and plants were sprayed with imidacloprid ( 50 mg/l ) to kill eggs . After 30 days , plants were evaluated for TYLCV infection based on symptoms and diagnostic PCR . Midguts of whiteflies that had had a 7-day AAP on TYLCV-infected tomato plant as well as TYLCV-unexposed control whiteflies were dissected with needles in DPBS and washed three times before use . The midguts were first fixed overnight with 2 . 5% glutaraldehyde in phosphate buffer ( 0 . 1M , pH7 . 0 ) and washed three times with phosphate buffer; then post-fixed with 1% OsO4 in phosphate buffer for 1 . 5h , washed , dehydrated in a graded series of ethanol ( 30% , 50% , 70% , 80% , 90% , 95% and 100% ) and embedded in Spurr resin . The specimen was sectioned using a LEICA EM UC7 Ultramicrotome . Sections were stained using uranyl acetate and alkaline lead citrate for 5 to 10 min , respectively , and observed in a Hitachi Model H-7650 TEM . Midguts of whiteflies were dissected freshly and fixed for 1 h with 4% paraformaldehyde ( PFA , Thermo Fisher , USA ) in phosphate-buffer saline ( DPBS; PH = 7 . 4 ) , then excess PFA was washed off with DPBS . Midguts were then permeabilized using 0 . 4% Triton X-100 in DPBS for 1 h and blocked using 3% BSA in DPBS for 2 h , followed by incubation with primary antibody ( overnight at 4°C ) . Midguts were subsequently incubated with secondary antibody and fluorescence labeled lectins for 2 h at room temperature , and finally fixed in 3% PFA with 0 . 05% glutaraldehyde in DPBS for 30 min . Monoclonal antibodies that recognize the coat proteins of TYLCV , TYLCCNV , and PalCuCNV were kindly provided by Professor Jian-Xiang Wu . The polyclonal antibodies against alpha-mannosidase 2 ( Man2 ) , Rab5 , Rab7 and TGN46 were produced by GenScript using synthetic peptides conjugated to KLH . The other antibodies and conjugated lectins were ERp57 polyclonal antibody ( PA5-29810; Invitrogen ) , Rab11 polyclonal antibody ( 2413; Cell Signaling Technology ) , goat anti-mouse IgG Alex 488 ( A11029; Invitrogen ) , goat anti-rabbit IgG Alex 647 ( A21245; Invitrogen ) , goat anti-rabbit IgG Alex 488 ( A11034; Invitrogen ) , lectin GS-II Alex 647 ( L32451; Invitrogen ) , lectin HPA Alex 647 ( L32454; Invitrogen ) , and lectin WGA Alex 647 ( W32466; Invitrogen ) . For confocal imaging , midguts were mounted in Fluoroshield Mounting Medium with DAPI ( Abcam , USA ) and viewed under LSM 780 ( ZEISS , Germany ) . ImageJ and JACoP were used for colocalization analysis with default parameters . For stochastic optical reconstruction microscopy , midguts were first immobilized on the bottom of Glass Bottom Cell Culture Dishes ( NEST , China ) and immersed in STORM imaging buffer containing cysteamine ( MEA ) and 2-mercaptoethanol ( 50 mM Tris-HCl pH = 8 . 0 , 10 mM NaCl , 10% Glucose , 5 mM MEA , 0 . 5% 2-mercaptoethanol , 560 μg/ml Glucose Oxidase , 34 μg/ml Catalase ) . Prior to STORM imaging , a strong laser was used to inactivate most fluorophores . STORM acquisition was then started with imaging cycles containing one frame of activation laser illumination ( 405 nm ) followed by eight frames of imaging laser illumination ( 561 nm , 657 nm ) . Typically , one STORM image containing more than 50 , 000 pictures was acquired per hour . These data were subsequently used for analysis of the locations of probes . HPA lectin was conjugated to NHS Mag Sepharose magnetic beads ( GE Healthcare , USA ) , following the manufacturer’s protocol . One g of whiteflies was ground in liquid nitrogen and dissolved in TBS ( 50 mM Tris , 150 mM NaCl , pH 7 . 5 ) with 0 . 4% NP-40 . Then , the solution was incubated overnight at 4°C with magnetic beads coupled with lectin HPA and washed three times with TBS . Finally , proteins on the magnetic beads were eluted using elution buffer ( 0 . 1 M glycine-HCl , pH 2 . 5 ) and subjected to mass spectrometry analysis at Shanghai Applied Protein Technology Co . , Ltd . Magnetic beads that were not coupled with lectin HPA were included as a control . The Mann-Whitney U test was used for comparisons of the relative abundance of virus in whitefly and the expression levels of genes . Goodness-of-fit test for independence was applied to comparisons of the transmission efficiency of TYLCV by whitefly . A P-value < 0 . 05 was considered as the threshold for significant difference . All the statistical analyses were performed with SPSS 20 . 0 ( SPSS Inc . , USA ) . | Many plant viruses are vectored by insects in a persistent circulative manner . In this process , the transport of virus from the gut lumen into the hemolymph of the vector is an important step . Identification of vector components involved in this transport process could lead to new strategies to combat virus spread . Here , we demonstrate that the early steps of endosomal trafficking are important for the transport of begomoviruses across epithelial cells in the whitefly midgut . After internalization by midgut epithelial cells , begomoviruses are transported directly from early endosomes to the basal plasma membrane in a fast recycling route . We also identified several host processes that are involved in begomovirus transmission , including vesicle transport and membrane fusion events that promote virus trafficking . To our knowledge , this is the first report on mechanisms governing the intracellular transport of begomoviruses in their insect vector . Our study provides new insights into the transmission mechanisms of circulative plant viruses and gives direction to future research into begomovirus transmission blocking strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"vesicles",
"cell",
"processes",
"microbiology",
"viral",
"structure",
"epithelial",
"cells",
"golgi",
"apparatus",
"plants",
"cellular",
"structures",
"and",
"organelles",
"endosomes",
"animal",
"cells",
"proteins",
"biological",
"tissue",
"tomatoes",
"cell",
"membranes",
"virions",
"fruits",
"biochemistry",
"lectins",
"eukaryota",
"cell",
"biology",
"secretory",
"pathway",
"anatomy",
"virology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2018 | Intracellular trafficking of begomoviruses in the midgut cells of their insect vector |
Noma is a spreading and fulminant disease believed to be native to Sub-Saharan Africa over the last decade and associated with low socioeconomic status of citizens of the region . Within this noma belt , most epidemiological reports regarding the disease have emanated from the north western region of Nigeria . However , our indigenous surgical mission encountered a substantial number of cases of noma and post-noma defects noteworthy of epidemiological representation across north central Nigeria . All noma cases encountered within the 8-year study period were included and divided based on clinical signs into acute and sequelae groups . Incidence estimation was based on acute/recently active cases and was calculated using the statistical method proposed by the WHO Oral Health Unit ( 1994 ) . Period prevalence of noma was calculated considering the population at risk in the zone . A total of 78 subjects were included in the study with age ranging from 2–75 years . Twelve subjects ( 15 . 4% ) presented with acute disease while 66 ( 84 . 6% ) had various forms of post-noma defects . The estimated incidence of noma in the north central zone was 8 . 3 per 100000 with a range of 4 . 1–17 . 9 per 100000 across various states . Period prevalence of noma which incorporated all cases seen within the study period was 1 . 6 per 100000 population at risk . Although noma may be more prevalent in the north western region of Nigeria , substantial number of cases occurs within the north central zone which calls for deliberate public awareness campaign on disease risk factors and prevention , and education of primary health-care providers .
Noma is a disease of the orofacial region that has been unanimously described as devastating , mutilating , destructive and debilitating due to its appearance and the nature of spread of the acute necrotizing lesion which runs fulminating courses . Alternatively known as Stomatitis gangrenosa or Cancrum oris , the aetiology of noma is infectious , yet unclear as regards the exact causative microorganism ( s ) [1 , 2] . Initially , a fuso-spirochetal microbial complex was implicated due to the higher level of these organisms in individuals with noma; however , this notion has been dispelled due to nonreproducibility in animal models following inoculation with these microorganisms under similar noma predisposing conditions [3] . Nonetheless , more recent breakthrough into the inquiry of noma microbiology has revealed a polymicrobial interaction between intraoral commensal organisms and extraoral opportunistic microbes as the most likely cause of the disease [2] . Although noma is almost exclusive to young children within ages two to six years; it has been shown to affect individuals across all age groups , progressing through unique clinical stages beginning at the reversible and seemingly inconsequential necrotizing gingivitis/oedema stages , to the grotesque gangrenous stage associated with extensive soft and hard tissue necrosis and a high mortality rate of 90% in untreated individuals [4–5] . In the presence of appropriate medical intervention at the latter stage of acute disease , scarring occurs–leaving sufferers with various forms of socially incapacitating facial defects which defines the chronic phase of the disease [5] . Despite being a disease first described over four centuries ago as affecting several world regions , noma is currently regarded as being exclusive to the tropics ( notably sub-Saharan Africa ) [6–7] , which is accredited to the preponderance of noma predisposing factors in the region . These factors include socioeconomic factors such as low standards of living , extreme poverty , poor sanitary conditions and close proximity of residence to livestock . Oral conditions such as poor oral hygiene and presence of simple gingivitis; systemic conditions like severe malnutrition , measles , malaria , tuberculosis , HIV infection , leukaemia , Non-Hodgkin’s lymphoma and cyclic neutropenia; and miscellaneous factors including low birth weight , improper weaning , birth position within the family and absence of mother as primary care giver [4 , 6 , 8–9] . Epidemiological research targeted at determining noma incidence and prevalence has been highlighted as a main feature of public health action programs against the disease . As determination of actual epidemiological parameters of noma is difficult due to high mortality associated with untreated disease , regional health data record inadequacies , remoteness of affected areas in addition to sufferers’ lack of access to primary health centres; current epidemiological data estimates a global incidence of 30 , 000–40 , 000 cases annually with seventy-five percent of these occurring in sub-Saharan Africa ( the noma belt ) [9–10] . Furthermore , in Nigeria ( particularly the north west and south west sub-regions ) , incidence rates between 0 . 8–6 . 4 per 1000 children have been reported in the last two decades mostly according to data provided by foreign non-governmental organizations or surgical missions [10–11] . Since 2010 , our indigenous surgical mission ( Cleft and Facial Deformity Foundation [CFDF] ) has embarked on organizing free intervention programs for individuals with orofacial conditions and deformities requiring urgent or elective surgical intervention in north central Nigeria–a region challenged with the dearth of craniofacial surgical expertise in secondary and tertiary health institutions . Although it is widely presupposed that the noma scourge is exclusive to northwest Nigeria as evidenced by the number of reports that have emanated from the region and the establishment of an health institution–Noma Children Hospital , solely concerned with treatment of acute stages of the disease and rehabilitation of survivors in the sub-region; cases of noma have also been encountered and successfully managed by our surgical mission across north central Nigeria within eight years . As epidemiological data is important for planning and prioritisation of service delivery as well as formulation of disease preventive strategies , we aim to provide an epidemiological report on noma disease in north central Nigeria by determining the incidence , prevalence , trend and risk factors for noma in the sub-region based on the health data records of noma cases encountered by our foundation over an eight years period spanning from 2010 to 2018 .
Comprehensive information was obtained from stored records of patients encountered in all surgical outreach programmes organized in north central Nigeria from June 2010 till September 2018 . All cases diagnosed as noma were included in this study , and this comprised both individuals with the acute disease or its sequelae . Since noma and orofacial cleft may share some similarities in clinical presentation , cases of the latter were excluded based on their congenital nature of occurrence and absence of significant morbidity associated with the deformity . Other head and neck or orofacial disease conditions were also excluded . The information obtained from the records included participants’ bio-data , year of encounter and facial location of defects at presentation . Distance between patients’ location of residence and the health institutions where the surgical outreaches were conducted was estimated for each participant . Other information included the number of siblings in the family ( <18 years ) , proximity of residence to livestock ( cattle , pigs , horses etc ) , primary caregiver around the time of disease onset , and history of visits to referral centres . Cases were also categorized into one of the five stages of the disease proposed by World Health Organization ( WHO ) –necrotizing gingivitis/beginner , oedema , gangrene , scar and sequelae; with the latter two stages indicative of long-standing or resolving disease [5] . In the custom of the Cleft and Facial Deformity Foundation Data Management Team ( CFDF-DMT ) , all health records are scrutinized at the end of every outreach program and variables perceived to be missing from a patient’s record are identified and eventually obtained from them at recall visits ( usually organized about two months following the surgical outreach program ) . For the purpose of this study , participants whose missing information could not be updated at the follow-up visits were excluded . This was necessary to ensure validity and reduce uncertainty of the research outcome . Data obtained from the study was analyzed using Statistical Package for Social Sciences ( SPSS ) version 23 . 0 ( IBM Corp Armonk , NY , USA ) . Descriptive statistics such as frequencies , mean and standard deviation were explored for quantitative and categorical variables as appropriate . The normality of the distribution was ascertained using the Shapiro-Wilk’s test . Difference between quantitative variables was determined using the Mann-Whitney U test while relationships between categorical variables were determined using the Pearson’s Chi-square test . The significance value of all statistical tests used were set to 5% ( p<0 . 05 ) . The prevalence of noma in the region was calculated by utilizing the total number of noma cases ( both acute and chronic phase ) seen within the study period as numerator and the population at risk as denominator . The population at risk considered only 45 . 7% of individuals residing below poverty line in the north central zone of Nigeria [12] . Incidence estimation analysis was done in line with the 1994 consultation report of the WHO Oral Health Unit using the Delphi method [13] . Only confirmed cases of acute noma ( ≤10years; beginner , oedema and gangrene ) or older sequelae cases who marked their sixth birthday within the study period ( giving due consideration to their year of encounter ) were included in the analysis . According to the method , estimating the total incidence ( I ) involves a two-step process , beginning with the determination of the total surviving cases . The number of surviving cases ( S ) is expressed as a function of the number of referred cases reaching our outreach or resident centres ( R ) and an approximation of the percentage of the total surviving cases that were referred ( χ ) which was approximated as 15% , considering the notion that about ‘one out of every five’ noma cases presenting to referral centres [13] and the fact that our programs were carried out quarterly . Thereafter , the total incidence ( I ) was extrapolated based on ‘S’ and the case survival rate of noma ( y; 10% ) . Ethical approval for the study was obtained from the Research Ethics review board of the International Craniofacial Academy . Prior notification of all outreach participants regarding the use of their health records and/or photographs for research purposes was done at each outreach event , with consents obtained . Records or images of participants/beneficiaries who refuse consent were never selected included or illustrated .
Age and sex variables of the noma cases seen were not normally distributed ( Shapiro-Wilk’s test , p<0 . 05 ) . Participants encountered were within ages 2–75 years with a majority of 43 . 6% ( n = 34 ) being above 30 years ( Table 1 ) . The average age of participants in this study was 29 . 6±18 . 84 years . Most individuals presenting with acute noma were between ages 2–10 years ( n = 10; 83 . 3% ) , with two subjects being adults aged 30 and 35 years; while approximately half of the sequelae cases which accounted for the most of the cases seen were above 30 years of age ( n = 33 , 50% ) [p = 0 . 001] . Analysis of their sex distribution revealed that males ( n = 42 , 53 . 8% ) were slightly more than females ( n = 36 , 46 . 2% ) , with a similar pattern of distribution obtained for both acute and sequelae cases [p = 0 . 735] . Most noma cases were observed in centres within Niger and Nasarawa states ( n = 48; 61 . 1% ) with only nine cases ( 11 . 5% ) recorded in Kogi state ( Table 1 ) . Four acute noma cases were encountered in Niger and Nasarawa states respectively ( denoting most of the cases ) , while subjects with noma sequelae defects were mostly encountered in Niger state ( n = 21; 31 . 8% ) . Since acute noma participants were mostly pre-schoolers or middle-age children , records based on occupation was only made to reflect if they were properly enrolled in school or not , which all participants in this category had no formal education at the time of encounter ( Table 1 ) . Comparatively in terms of economic status , most participants with noma sequelae ( n = 28 , 42 . 4% ) were not productively engaged , while 18 . 2% had careers centred on agriculture ( farming , fishing or cattle rearing ) . Of the twelve ( 12 ) acute noma cases recorded , 91 . 7% exhibited features of gangrenous stage of the disease ( n = 11 ) , while only one subject was noted to have presented with facial swelling and necrotic ulcerations of the mucosal lining of the upper lip and cheek which are consistent with the oedema stage of noma . Thirteen of the 66 participants with post-noma defects ( 19 . 7% ) had nascent scarring indicative of recent active disease , with 53 ( 80 . 3% ) showing clinical signs that are indicative of stage five ( full blown sequelae ) of the noma disease spectrum . Regarding the facial locations of the noma defect , 50% of all cases ( n = 39 ) had deformities involving the nose while 28 . 2% ( n = 22 ) and 32 . 1% ( n = 25 ) had lesions that affected their right and left cheek respectively . Of both lips , the upper lip was mostly affected ( n = 30 , 38 . 5% > n = 13 , 16 . 7% ) , and the occurrence of trismus among noma sequelae sufferers was 13 . 6% ( n = 9 ) . Within the study period , acute noma cases were first encountered in 2012 , with a steady rate of occurrence from 2013 to 2016 , and increase in the number of cases encountered in 2017 ( n = 5 , 41 . 7% ) . However , no cases of acute noma were seen within the study period of 2018 ( Fig 1 ) . Participants with post-noma defects were seen annually throughout the study period , with cases increasing from two ( 3 . 0% ) to fourteen ( 21 . 2% ) from 2010 to 2011 . Alternating decline and increase of noma sequelae cases were thereafter observed from 2011 to 2018 ( Fig 1 ) . The highest number of noma cases seen in a single year was 14 ( 17 . 9% ) , which occurred in 2011 and 2017 . While all the participants encountered in 2011 had post-noma defects , five out of the fourteen cases in 2017 had acute noma lesions ( 35 . 7% ) . Table 2 shows the assortment of all noma participants ( acute and chronic ) presenting to the outreach referral centres within different parts of the zone . Pertinent information collected regarding risk factors associated with noma included the number of siblings in the family , being raised by extended family members ( especially grandparents ) , proximity of household residence to livestock and distance between residence and location of health outreach facility . The number of siblings of participants ( < 18 years ) ranged from 3 to 18 in total with an average of 8 . 6 ± 5 . 06 . Furthermore , 85 . 9% ( n = 67 ) answered positively that they lived in close proximity to livestock or even reared them while only 19 . 2% ( n = 15 ) admitted to residing with extended relatives around the time of onset of noma disease . The mean distance between patients’ residence and location of the health facility used for the surgical outreach was 124 . 8 ± 96 . 713km , with 41 ( 52 . 6% ) participants residing at locations 30 to 100 km from the referral centres . Twenty-eight ( 35 . 9% ) individuals live between 101 to 300km away from the host secondary health facilities while 5 ( 6 . 4% ) and 4 ( 5 . 1% ) participants dwell in areas <30km and >300 km from the centres respectively . Participants’ records further revealed that 60 ( 76 . 9% ) subjects had never visited a health referral centre in the past and cited our indigenous surgical mission as the first centre of presentation since the disease onset . The total estimated incidence of noma in the north central region of Nigeria between 2010 and 2018 is 8 . 3 per 100 , 000 population , with approximately 109 new cases diagnosed annually ( 17–42 cases across the various captured states , within the geopolitical zone ) . Noma incidence was highest in Nasarawa state with a rate of 17 . 9 cases per 100 , 000 population and lowest in Kogi state with an incidence of 4 . 1 cases per 100 , 000 population . The incidence extrapolated for Niger state and the Federal capital territory ( FCT ) was 5 . 1 and 14 . 2 per 100 , 000 respectively . The period prevalence of noma in this study is 1 . 6 cases per 100 , 000 population at risk , with calculated sex occurrence rates being 1 . 7 per 100000 for males and 1 . 5 per 100000 for females . Noma prevalence was highest in the Federal capital territory ( 3 . 3 per 100 , 000 population at risk ) , with proportions ranging between 0 . 6–1 . 4 per 100 , 000 obtained in Kogi ( 0 . 6 ) , Nasarawa ( 1 . 3 ) and Niger ( 1 . 4 ) states .
As a means of overcoming the challenges posed by the paucity of noma epidemiological data , nongovernmental organizations were one of relevant stakeholders saddled with the responsibility of reporting cases of noma sufferers and survivors encountered , in a bit to raise awareness on disease occurrence across their various stations or referral centres within the noma belt region [14] . Over the study period , our volunteer-based surgical mission discovered a substantial number of noma cases noteworthy of epidemiologic representation in north central Nigeria , which would allow for adequate characterization of the disease burden in this region–an area resident to the third-highest number of citizens living below poverty line in Nigeria [12] . In addition , our study shifts major attention from the recent norm of conducting noma epidemiological surveys in the north western region of Nigeria ( Sokoto state in particular ) over the last decade to the north central region comprising six member states ( Benue , Kogi , Kwara , Nasarawa , Niger , Plateau ) and the Federal Capital Territory . Attempts at estimating noma incidence commenced towards the end of the 20th century . Barmes et al [15] first reported case incidence extrapolations of noma from Niger , Nigeria and Senegal , which followed in 1998 with the world health report by Bourgeois and Leclerq–initially estimating an incidence of 140 , 000 cases worldwide from interviews with health workers in noma prevalent areas [16] . Fieger et al [10] from the most recent report on noma incidence in north west Nigeria , estimated an incidence of 25 , 600 cases in developing countries bordering the Sahara Desert ( the noma belt of the world ) , and a global incidence of 30 , 000–40 , 000 cases . In like manner , our study estimates a noma incidence of 8 . 3 per 100000 in north central Nigeria from 2010–2018 , with a range of 4 . 1–17 . 9 per 100000 observed across different states within the geopolitical zone . This estimate is approximately eighty folds less than the calculated incidence of noma reported by Fieger et al [10] from 378 noma patients encountered between 1996–2001 in Sokoto , Nigeria . In the latter study ( based on a multiple logistic regression model of deducing unknown noma incidence from available incidence data of orofacial cleft within the region ) , the estimated incidence was 6 . 4 per 1000 with values varying between 4 . 4 and 8 . 5 per 1000 in individuals aged 10–30 years . Our lower incidence estimate may be clearly attributed to the wide variation in poverty indices of the north western and north central region of the country over several years , with the north western region serially recording the highest number of individuals living below poverty line in the country ( >80 . 0% ) [12] . By inference therefore , the north central zone may have less residents with severe malnutrition , unsafe drinking water , poorer sanitation practices and limited access to proper healthcare when compared to the north western sub-region . A supporting reason for the wide variation observed as compared to the reports of Fieger et al [10] may be the inclusion of noma cases above 10 years of age and possible noma sequelae cases in the sample utilized for the incidence estimation in the latter which implies probable over-estimation in the incidence values extrapolated for the north west region of Nigeria . Our calculated incidence was also lower than the values obtained by Denloye et al [17] in Ibadan , south west Nigeria , where an incidence rate of 7 . 0 per 1000 cases was reported in individuals within ages 1 to 12 years from 1986–2000 . This finding may be attributed to the disparate methodology of incidence calculation in both studies , as the forty-five noma cases reviewed by Denloye et al [17] were used against the total number of children that presented to the referral centre within the study period . In further comparison of our findings with previous reports from other countries within the noma belt of the world , our estimated incidence was lower than the case incidence reported from Niger Republic ( 1 . 34 per 1000 ) and Senegal ( 0 . 7–1 . 2 per 1000 ) by Barmes et al [15] among children aged 0–6 years . However , this comparison may be flawed since at the time of data extrapolation , a seemly unrealistic mortality rate of 70% was utilized for the incidence estimation in their study; this observation was also highlighted by the reports of Fieger et al [10] . Subsequently , incidence estimates adjusted to a more accurate noma mortality rate of 90% resulted in incidence estimates between 1 . 2 to 4 . 2 cases per million in Dakar , Senegal among children aged 0–9 years of age [18] , which ranks lower than the calculated incidence in our study . The geographic distribution of noma is commonly represented figuratively on world maps by the WHO and its Regional offices , and titled “Noma in the world” [13] . These maps , which were first published in 1994 , were made to depict reported cases of the disease across various parts of the world based on available data at the time of publication; hence , providing a diagrammatic panorama of the current noma situation in the world . The last update of these maps ( published around 2009 ) [19] showed then recent noma case observations in 67 . 9% of African countries , no recent reported cases in two countries ( DR Congo and Morocco ) and sixteen “noma free” countries ( Swaziland , Lesotho , Liberia , Sierra Leone , Guinea Bissau , Congo , Gabon , Libya , Tunisia , Equatorial Guinea , Burundi , Rwanda , Eritrea , Comoros , Mauritius and Western Sahara ) within the African continent . Other affected continents included Asia ( India , Pakistan , Myanmar ) , South America ( Colombia , Guyana , Suriname , Argentina , Paraguay and Uruguay ) , with sporadic reports in Oceania , Europe and North America . With no recent map publication for over a decade , there is no current ‘snapshot’ or precise description of noma case observations worldwide , as well as no performance indicators of preventive strategies targeted at disease prevalent areas . Therefore , an update of these ‘noma maps’ by relevant monitoring stakeholders based on recent data/reports from experts within the last decade is urgently required in Africa and indeed globally to allow for knowledge acquisition on current ‘high risk’ areas and concomitant implementation of primary and secondary preventive approaches in these regions . The pattern of noma/post-noma soft tissue defects in north central Nigeria includes varying degree of deformities involving the nose , upper lip , left cheek , right cheek and lower lip in decreasing order of occurrence . Our finding corroborates the pattern of noma presentation observed in Dakar Senegal in which the upper lip then cheek were quoted as the two most common sites affected by the noma defect; although , cases involving the nose were not cited in their report [18] . In contrast , the site distribution of defects in our study varies from the recent observations of Adeniyi and Awosan in Sokoto , north west Nigeria where lesions involving the cheeks were mostly seen , followed by upper and lower lips with the nose being the least affected site of soft tissue defects [20] . Farley et al [9] in a case-control study involving 74 cases and 222 controls in north west Nigeria , associated “caretaker” ( i . e third party carer ) as a factor that may influence the risk of developing noma . This was supported by Adeola et al [21] as reported in a case series involving five subjects managed for acute phase of noma in north west Nigeria . In that study , they asserted there was increasing number of noma cases in the region associated with lack of direct maternal care after children were weaned . This observation was not common in our study where only 19 . 2% of noma patients were being cared for by extended relatives around the time of disease onset . Although this proportion may be considerable , it was not significant enough to corroborate the observation of the earlier authors . Hence , it may require further studies to determine the plausibility of this assertion . Another important observation is the travel distance to access health care for patients in the north central sub-region . In this study , subjects had to travel about 125km on the average to access care from our treatment teams at the secondary health centres . In fact , 76 . 9% of them had not presented to any referral centre previously during the course of active disease or after its remission . Our experience in the region attributes this observation to factors such as subject’s preference for self-medication or traditional medical alternatives , lack of primary/secondary health centres within close proximity to residence , presence of primary centres but lack of required expertise and facilities for treatment , available facilities and expertise within state of residence but lack of finances to offset the cost of treatment and additional cost of transportation . Since the extrapolations in this study were based on data records obtained from four states within the sub-region , the non-availability of records from Plateau , Benue and Kwara states represents a limitation to our incidence estimation; although , most areas are largely sub-urban/urban with tertiary health institutions in each of the states mentioned . Also , the possibility that some of the subjects may have developed the disease in these neighbouring north central states previously and relocated to the state of encounter prior to the outreach cannot be totally ruled out . Another limitation is the inability to explore in details , some major risk factors for noma due to the retrospective nature of the study design . Furthermore , the use of non-probability ( convenience ) sampling methods to arrive at the sample size employed for the analysis of associated risk factors for noma in this study may have introduced selection bias . Considering that our study sought out to primarily determine the estimated incidence of noma in north central Nigeria , the results that have emanated are specific to characterizing the burden of noma in this zone . However , the statistical methods for the calculations done were adapted from the WHO Oral Health Unit and could as well be applied to other regions of the country or in other locations within the noma belt .
The estimated incidence of noma in north central Nigeria is 8 . 3 cases per 100000 population . Although the noma scourge is deemed prevalent in north west Nigeria and Sokoto state in particular , substantial number of cases is being encountered in the north central zone . Hence , efforts should be intensified in terms of public awareness , establishment of new primary health centres in deficient councils/wards , and education of community health workers in existing primary health care centres on disease identification ( possibly primary care ) in order to facilitate presentation of sufferers to appropriate referral centres within the north central zone . With a prevalence of 1 . 6 per 100000 population at risk and majority living with post-noma defects , it is clear that attention to surgical rehabilitation in the region is also suboptimal . It is therefore imperative in the absence of any health facility solely dedicated to the management and rehabilitation of noma patients in the region ( unlike the northwest ) ; that existing secondary health centres and nongovernmental organizations in the zone be better equipped to mitigate the disease burden and provide standard care for noma cases and survivors , especially as the poverty index of the zone and country is increasing . We further recommend that maps denoting noma occurrence in Africa and globally be updated according to recent available data so as to reflect current disease distribution and enable targeted preventive strategies in identified ‘high risk’ nations . | Noma , a devouring disease of the orofacial complex , is commonly associated with poverty and impoverished regions of the world especially Sub-Saharan Africa termed the noma belt region of the world . With more reports advocating for full inclusion of noma in the WHO Neglected Tropical Diseases ( NTDs ) program , the apparent neglect of the disease in north central Nigeria compared to other sub-regions is worrisome as the disease burden in the sub-region has not been reported till date . In this light , a retrospective , cross-sectional survey was conducted to provide epidemiological representation to the cases encountered within an eight-year period by the Cleft and Facial Deformity Foundation ( CFDF ) , an indigenous surgical mission . The incidence of noma was estimated using methods recommended by WHO while the period prevalence was calculated considering the population living below poverty line in the sub-region . This study extrapolates an incidence of 8 . 3 cases per 100000 and a period prevalence of 1 . 6 per 100000 persons at risk . Notable is the finding that most individuals encountered were above thirty years of age and suffered varying degree of facial disfigurement consequent to acute noma disease experienced in their childhood/early adolescence . Therefore , we advocate public awareness on the disease risk factors and prevention within the sub-region as well as training of primary health personnel on disease identification , primary care and nearest referral centres . We also identify the need to bolster the efforts of existing health facilities and indigenous surgical missions in the management and rehabilitation of noma cases and survivors . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusions",
"and",
"recommendations"
] | [
"medicine",
"and",
"health",
"sciences",
"face",
"niger",
"tropical",
"diseases",
"geographical",
"locations",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"health",
"care",
"bacterial",
"diseases",
"neglected",
"tropical",
"diseases",
"africa",
"veterinary",
"science",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"epidemiology",
"noma",
"head",
"nigeria",
"people",
"and",
"places",
"socioeconomic",
"aspects",
"of",
"health",
"anatomy",
"biology",
"and",
"life",
"sciences"
] | 2019 | Estimated incidence and Prevalence of noma in north central Nigeria, 2010–2018: A retrospective study |
Cohesin is crucial for proper chromosome segregation but also regulates gene transcription and organism development by poorly understood mechanisms . Using genome-wide assays in Drosophila developing wings and cultured cells , we find that cohesin functionally interacts with Polycomb group ( PcG ) silencing proteins at both silenced and active genes . Cohesin unexpectedly facilitates binding of Polycomb Repressive Complex 1 ( PRC1 ) to many active genes , but their binding is mutually antagonistic at silenced genes . PRC1 depletion decreases phosphorylated RNA polymerase II and mRNA at many active genes but increases them at silenced genes . Depletion of cohesin reduces long-range interactions between Polycomb Response Elements in the invected-engrailed gene complex where it represses transcription . These studies reveal a previously unrecognized role for PRC1 in facilitating productive gene transcription and provide new insights into how cohesin and PRC1 control development .
The cohesin complex that encircles DNA is named for its role in mediating sister chromatid cohesion [1] . Cohesin has four subunits: the Smc1 and Smc3 structural maintenance of chromosomes proteins , Rad21 , a kleisin protein , and Stromalin ( SA ) . Cohesin is loaded onto chromosomes by the kollerin complex containing the Nipped-B ( NIPBL , Scc2 ) and Mau-2 ( Scc4 ) proteins , and removed by the releasin complex containing Pds5 and Wapl . Minor alterations in cohesin function disrupt development without affecting sister chromatid cohesion or chromosome segregation [2] . In humans , dominant loss-of-function mutations in the NIPBL gene encoding a kollerin subunit , and dominant missense mutations in the Smc1 or Smc3 cohesin subunits cause Cornelia de Lange syndrome ( CdLS ) [3] . CdLS is characterized by poor growth , diverse structural abnormalities and intellectual impairment . Current data argue that small changes in cohesin function disrupt development because cohesin binds and regulates many genes important for growth and development [2] , . Genetic and molecular evidence suggests that cohesin also functionally interacts with Polycomb group ( PcG ) epigenetic silencing proteins during development . There are two major PcG protein complexes: Polycomb Repressive Complex 2 ( PRC2 ) , whose Enhancer of zeste [E ( z ) ] subunit performs histone 3 lysine 27 trimethylation ( H3K27me3 ) , and PRC1 , whose Polycomb ( Pc ) subunit binds the H3K27me3 histone modification [5]–[8] . The Drosophila Rad21 cohesin subunit is encoded by verthandi ( vtd ) . Nipped-B kollerin and vtd cohesin subunit mutations suppress the body segment transformations of Pc mutant flies , and a wapl releasin mutation that stabilizes cohesin binding causes phenotypes similar to Pc mutations [9]–[11] . The in vivo findings argue that cohesin antagonizes PcG silencing at certain genes , and consistent with this idea , genome-wide mapping in Drosophila cells revealed that cohesin and Nipped-B preferentially bind promoters of active genes , and are usually excluded from PcG-silenced genes [12] . The few exceptional genes that bind cohesin and also have the PRC2-generated H3K27me3 mark are not fully silenced , and show unusually large increases in expression upon cohesin depletion [13] . Cohesin biochemically interacts with PRC1 , suggesting that these two complexes might directly control each other's activities [14] . To clarify the functional relationships between cohesin and PcG complexes , we mapped their binding genome-wide in developing Drosophila wing and cultured nervous system cells , and measured the effects of cohesin on PRC1 binding , and vice versa . We also measured the genome-wide effects of PRC1 depletion on Pol II occupancy and mRNA levels . These studies revealed functional interactions between cohesin and PRC1 that give rise to their co-regulation of active genes , and antagonism to each other on expression of PcG-silenced genes . They uncovered an unexpected role for PRC1 in controlling RNA polymerase II ( Pol II ) function at active genes that provides new insights into how cohesin and PcG proteins regulate transcription .
The genomic data from cultured cells and genetic interactions between cohesin and PcG mutations suggest that cohesin and PcG proteins functionally interact to control gene expression and development . To test this idea in a developing tissue , we used genomic chromatin immunoprecipitation ( ChIP-chip ) to map Rad21 , Nipped-B , RNA polymerase II ( Pol II ) , and PRC2-generated histone 3 lysine 27 trimethylation ( H3K27me3 ) in late 3rd instar Drosophila wing imaginal discs . Genetic data shows that both cohesin and PcG proteins regulate gene expression and development in wing discs ( e . g . [15]–[17] ) . Two biological replicates were used for each ChIP-chip experiment , and enrichment was calculated genome-wide using the model-based MAT algorithm [18] , [19] . We found that the cohesin binding patterns seen in Drosophila cell lines [12] hold true in developing wing: ( a ) cohesin ( Rad21 ) co-localizes with Nipped-B ( genome-wide correlation r = 0 . 91 ) , ( b ) cohesin and Nipped-B preferentially bind active genes occupied by Pol II ( r = 0 . 63 ) , and ( c ) cohesin and Nipped-B are largely absent from silenced genes with PRC2-generated H3K27me3 ( r = 0 . 24 ) . The genome-wide correlation coefficients are summarized in Table S1 . These and the other genome-wide ChIP and expression assays performed for this study ( see below ) are listed in Table S2 . Table S3 lists all annotated genes , and indicates whether or not they bind each of the factors that were mapped by ChIP-chip in wing discs and cultured cells ( see below ) . Table S3 also gives the mRNA expression values determined by microarray . The correlation between cohesin and H3K27me3 is higher in wing discs than in cultured cells . It is likely that sites of false overlap arise from the mixture of different cell types in the wing disc , in which a gene can be silent in one compartment and active in another . For example , the invected-engrailed ( inv-en ) gene complex , encoding two homeobox transcription factors that confer posterior fate , is expressed in the posterior wing compartment and PcG-silenced in the anterior ( e . g . [16] , [20] ) . The inv-en complex is of particular interest because it is one of the rare examples of cohesin-H3K27me3 overlap in BG3 cells , where its expression is highly sensitive to cohesin dosage [13] . We thus mapped cohesin ( Rad21 ) , H3K27me3 , and the PRC1 subunit Polycomb ( Pc ) in the posterior and anterior wing disc separately to determine if inv-en might be regulated by both cohesin and PcG proteins in either the active or silenced state . We used a transgene that expresses red fluorescent protein ( RFP ) only in the posterior compartment [21] as a guide to slice the discs into anterior and posterior portions ( Figure 1A ) . We found that inv-en binds cohesin primarily in the posterior disc , where it is expressed , and is marked by H3K27me3 primarily in the anterior disc , where it is silenced ( Figure 1A ) . The low levels of cohesin detected in anterior chromatin , and low levels of H3K27me3 seen in posterior chromatin , likely reflect imperfect dissection ( 10–20% cross-contamination by visual inspection and by measuring inv and en RNA levels , Figure S1 ) . This low level of contamination is unavoidable , given the difficulty of the dissection by hand of discs that are some 250 by 350 microns in size . Tissue dissociation and FACS sorting did not provide sufficient viable cells for chromatin preparation . Nonetheless , the data clearly demonstrate an inverse relationship between H3K27me3 and cohesin at the inv-en complex between the anterior and posterior compartments . The genome-wide correlation between Rad21 and H3K27me3 in the posterior wing disc ( r = 0 . 11 ) is significantly lower than for whole discs ( r = 0 . 24 ) , further confirming that some sites of Rad21-H3K27me3 overlap seen with whole discs reflect tissue heterogeneity , and that H3K27me3 and cohesin have an inverse relationship genome-wide in wing discs as in cultured cells ( Table S1 ) . We unexpectedly found that the Pc PRC1 subunit occupies inv-en and flanking active genes at virtually equal levels in anterior and posterior wing discs ( Figure 1A ) . Even more surprising , by measuring the mRNA levels for some 13 , 000 genes in whole wing discs ( Table S3 ) and comparing to the genomic ChIP data , we found that Pc associates with a large portion of active genes , including ubiquitously expressed genes such as the Act5C actin gene , diminutive ( dm , the Drosophila myc gene ) and some ribosomal protein genes ( Figure 1B ) . Among active genes , Pc preferentially occupies those that bind cohesin ( Figure 1D ) . In the posterior disc , Pc and Rad21 binding exhibit a genome-wide correlation of 0 . 75 ( Table S1 ) , and Pc is present at 90% of cohesin-bound genes ( Figure 1D ) . Conversely , 76% of Pc sites exhibit cohesin binding . The extensive overlap of cohesin and Pc at active genes was unexpected , given that PRC1 is generally thought to associate primarily with PcG-silenced genes . For instance , Pc was previously mapped in Drosophila cell lines with a different antibody ( denoted Pc-VP; [22] ) than the one used here ( denoted Pc-RJ; [23] ) , and was generally found at silent genes marked by H3K27me3 [24] , [25] . Both Pc-VP and Pc-RJ were made using the same Pc fragment as antigen ( residues 191–354 ) and antigen-affinity purified . The Pc-VP antibody was validated for ChIP-chip in the initial studies , and re-validated by the modENCODE project , including RNAi depletion experiments . Two laboratories independently verified the Pc-RJ antibody by RNAi , western blots and ChIP [13] , [23] . Figure S2 confirms that the Pc-RJ antibody detects a single protein of the expected size that is reduced in cultured cells subjected to Pc RNAi treatment . The core PRC1 complex consists of Pc , Polyhomeotic ( Ph ) , Posterior sex combs ( Psc ) and Sex combs extra ( Sce/dRing ) . We thus further mapped PRC1 in whole wing discs using the Pc-VP Pc antibody and validated antibodies against the Polyhomeotic ( Ph ) and Posterior sex combs ( Psc ) PRC1 core subunits [22] . These antibodies were previously validated by westerns , in vivo expression of fusion proteins , and ChIP-chip [22] , [24]–[26] . Figure S2 shows that the Ph antibody recognizes one major band of the correct size that is reduced upon treatment of cultured cells with Ph RNAi . These experiments demonstrate that the PRC1 complex is present at active cohesin-binding genes in wing discs . Ph ChIP gave a nearly identical pattern to Pc-RJ , with a genome-wide correlation with Rad21 of 0 . 82 in whole discs ( Figure 1B , C , E; Table S1; Figure S3 ) . Thus two validated antibodies against different PRC1 subunits show essentially the same pattern . It is very unlikely that both cross-react with a non-PRC1 protein that co-localizes with cohesin by chance . Moreover , Psc was also detected at many cohesin-binding active genes , although the signals are noticeably lower than at silenced genes ( Figure 1B , C , E; Figure S3 ) . We cannot distinguish if the lower Psc signals reflect reduced epitope accessibility , or if Psc is present in a smaller fraction of the PRC1 complexes at active genes . For instance , Su ( z ) 2 , a Psc homolog encoded by a neighboring gene , may substitute for Psc in a much higher fraction of PRC1 complexes at active genes . Figure S3 shows plots of the ChIP enrichment at all microarray features for each of the PRC1 subunits against each other over a 400 kb region on chromosome 2L that includes several active genes and the PcG-silenced dpp gene , along with the genome-wide correlation coefficient for each comparison . Ph shows a high genome-wide correlation with Pc-RJ ( 0 . 85 ) , while Pc-VP shows a lower correlation , with the plots showing a clear separation of active and silenced genes . Psc correlates with Ph at both silenced and active genes , with a distinct separation into silenced genes with high Psc and active genes with low Psc . Figure S4 shows a high resolution view of the diminutive ( dm , myc ) gene to show the close similarities in the ChIP patterns of three PRC1 subunits ( Pc , Ph , Psc ) at a constitutively-active cohesin-binding gene . Figure S5 shows that additional independent anti-Pc and anti-Ph antibodies [27]–[29] also detect PRC1 at active cohesin-binding genes by ChIP-qPCR . We thus conclude that PRC1 is present at most active cohesin-binding genes . As in cultured cells , the Pc-VP antibody detected Pc binding primarily at sites of H3K27me3 in wing discs , such as inv-en ( Figure 1C , E ) . We do not know why Pc-VP , in contrast to several other PRC1 antibodies , does not detect PRC1 at active genes , but one clear possibility is that the Pc epitope recognized by Pc-VP is masked at active cohesin-bound loci . This idea is consistent with the direct interaction between cohesin and PRC1 detected by purification of biotin-tagged PRC1 [14] . The cohesin-PRC1 interaction was characterized using different cohesin and PRC1 antibodies than those used in our studies . With the antibodies we used for genomic ChIP , immunoprecipitation of Rad21 from soluble nuclear extracts treated with DNase I co-precipitated Pc and Ph , confirming that cohesin and PRC1 interact ( Figure S2 ) . These results also provide further validation of the specificity of the Pc-RJ and Ph antibodies . The extensive overlap of cohesin and PRC1 at active genes in wing discs and the direct interaction between cohesin and PRC1 suggested that they might influence each other's binding . The presence of multiple cell types in wing discs creates ambiguities in interpreting ChIP signals at genes that are not active or silent in all cells . It is also difficult to reduce cohesin and PRC1 levels in vivo by more than 50% without causing lethality or substantially altering development . Homozygous cohesin and PcG mutants are early lethals , and many putatively tissue-specific RNAi drivers cause lethality even without large reductions in cohesin [30] . We thus chose to examine the effects of cohesin on PRC1 binding and vice versa in cultured ML-DmBG3 ( BG3 ) cells derived from 3rd instar central nervous system as a more homogenous cell population , and in which cohesin or PRC1 subunits can be easily reduced by at least 80% using RNAi without causing lethality [13] . We mapped Pc and Rad21 binding genome-wide in BG3 cells using an antibody ( Pc-RJ ) that detects PRC1 at active genes . Cohesin binds primarily at active genes with promoter-proximal paused RNA polymerase II ( Pol II ) in BG3 cells , and controls the transition of paused Pol II to elongation [12] , [13] , [31] , [32] . There is extensive overlap between cohesin and Pc in BG3 cells , although lower than seen in wing discs , with 72% of cohesin-bound genes binding Pc ( Figure 2A ) . We conclude , therefore , that PRC1 also binds most active cohesin-binding genes in BG3 cells , including several highly-expressed constitutively-active genes such as some ribosomal protein genes , Act5C , and dm/myc ( Figure 3 ) . We tested if cohesin and PRC1 influence each other's binding by performing genomic ChIP of Rad21 after Ph RNAi depletion , and Pc ChIP after Rad21 RNAi depletion , with two biological replicates for each RNAi treatment . Depletion of cohesin by 80% in BG3 cells does not alter chromosome segregation , cell morphology or viability , and modestly decreases proliferation [13] , [31] . This depletion reduces cohesin-binding by 70% or more at genes with high levels of cohesin , and large changes in expression of several genes that bind cohesin [13] , [31] , [32] . Figure S6 shows that cohesin depletion increases the fraction of cells in G2 without a substantial effect on the proportion of cells in S phase . Prior analysis , including genome-wide mRNA measurements , suggests that the G2 delay reflects decreased diminutive ( dm , myc ) gene expression and cell differentiation , and not mitotic defects [13] . There are minimal effects on sister chromatid cohesion , and no detectable effects on chromosome segregation . There is a modest 1 . 5-fold increase in cyclin B mRNA and modest decrease in expression of genes involved in spindle formation and elongation , indicating a delay prior to entry into mitosis . There are no changes in expression of cell cycle checkpoint or apoptosis genes . Consistent with decreased dm/myc expression , the most significant down-regulated gene ontology is protein synthesis , and the top up-regulated gene ontology is development , suggesting a change in cell differentiation . Both the up- and down-regulated genes are highly enriched for cohesin-binding genes , indicating that a large proportion of the affected genes are directly regulated by cohesin , and that most expression changes are not attributable to changes in cell physiology . Although the levels of mRNA produced by most genes that do not bind cohesin do not change upon cohesin depletion , direct transcription measurements show that transcription initiation subtly decreases at most of them , which likely reflects decreased levels of Dm/Myc or other general transcription factors [32] . Similar to cohesin depletion , PRC1 depletion also increases the fraction of cells in G2 without decreasing the proportion of S phase cells ( Figure S6 ) . Depletion of Ph , similar to Pc depletion [13] does not reduce viability , but causes BG3 cells to form more distinct colonies , with longer cellular processes , and cessation of proliferation after six to seven days of treatment . We thus examined the effects of Ph on Rad21 binding after five days of RNAi treatment , when Ph is reduced by at least 80% ( Figure S2 ) and cells are still dividing , to minimize binding changes that might be caused by cell differentiation . Changes in Pc binding upon Rad21 depletion were measured by two methods ( Figure S7 ) , both of which showed that Rad21 depletion reduces Pc binding to active genes . In the first method , we integrated the Pc ChIP MAT scores for all microarray features within a gene body in the control and Rad21-depleted cells , and then calculated the difference in the total ChIP signal for each gene between the depleted and control cells ( Figure S7 , method 1 ) . By comparing these differences to the control mRNA level for each gene ( Table S3 ) , we found that Pc binding decreases at many active genes upon Rad21 depletion ( Figure 2B ) . We also measured the change in Pc ChIP enrichment after Rad21 depletion at each individual microarray feature ( ΔPc ) across the genome , and then mapped all locations where ΔPc was greater or smaller than the median genome-wide ΔPc by at least two standard deviations for at least three microarray features in a row against all annotated genes ( Figure S7 , method 2 ) . These intervals are shown in the Δ tracks in Figure 3 and Figure S8 . We found that 50% of active cohesin-binding genes show a Pc decrease , compared to only 10% of the genes that don't bind cohesin , and that Pc increases are 10-fold less frequent than increases at cohesin-binding genes ( Figure 2C ) . The finding that cohesin depletion frequently decreases but rarely increases PRC1 binding to active cohesin-binding genes argues that most of the changes in PRC1 binding are a direct consequence of cohesin reduction at the cohesin-binding gene , and not caused indirectly by cellular differentiation . Differentiation would be expected to have more selective , gene-specific effects , and would also be more likely to increase PRC1 at some active genes . Together with the direct interaction between cohesin and PRC1 ( Strübbe et al . , 2011; Figure S2 ) , the decrease in PRC1 levels at many cohesin-binding genes argues that cohesin recruits or stabilizes PRC1 binding at active genes . Because cohesin depletion alters transcription of many genes , we cannot rule out the possibility that the PRC1 decreases upon cohesin depletion are caused in part by transcriptional changes . We note , however , that the effects of cohesin reduction on PRC1 binding do not fully mirror the effects on Pol II occupancy . Although cohesin depletion frequently reduces both total Pol II ( Rpb3 ) and Pc at cohesin-binding genes , the genome-wide correlation between the change in Pc ( ΔPc ) and change in Rpb3 ( ΔRpb3; [32] ) is modest ( r = 0 . 29 ) . In addition to the effects of cohesin depletion on PRC1 binding , we found that PRC1 depletion affected cohesin binding at active genes . Ph depletion increased Rad21 signals at nearly a third of active cohesin-binding genes ( Figure 2B , Figure 2C , Figure 3 , Figure S8 ) . Cohesin increases were twice as frequent as decreases ( Figure 2C ) , and increases were generally greater in magnitude than decreases ( Figure 2B ) . These findings suggest that although cohesin facilitates PRC1 binding , PRC1 negatively influences cohesin binding at many active genes . We cannot rule out , however , the possibility that some increases in cohesin signals are caused by greater epitope accessibility in the absence of PRC1 , or increased Pol II levels in the gene body ( see below ) . In contrast to active genes , cohesin and PRC1 binding are mutually antagonistic at PcG-silenced genes that are marked by H3K27me3 and bind Pc . Rad21 binding increases at 28% of these genes upon Ph depletion , and Pc binding increases at 60% of them upon Rad21 depletion ( Figure 2B , C ) . The Rad21 increases upon Ph depletion , and the Pc increases upon Rad21 depletion occurred at all the genes in the Antennapedia and bithorax HOX gene complexes ( ANT-C , BX-C ) . The changes at the Sex combs reduced ( Scr ) and Antennapedia ( Antp ) genes in the ANT-C are shown in Figure 3 . The increased cohesin binding is not caused by higher cohesin expression , because Ph depletion slightly reduces cohesin subunit mRNA levels , as determined from a genome-wide mRNA analysis described below . The increase in Rad21 at silenced genes , however , correlates with an increase in Pol II and mRNA for the silenced genes ( see below ) , and thus increased cohesin binding to silenced genes could be caused by increased transcription . The increased Pc levels at silenced genes upon Rad21 depletion are unlikely to reflect higher PRC1 expression . Rad21 depletion increases the Psc and Su ( z ) 2 mRNAs less than 2-fold , and has no significant effect on the expression of other PRC1 subunits [13] . Also , as described above , Rad21 depletion decreases PRC1 levels at active genes . The increase in Pc at silenced genes upon Rad21 depletion does not reflect a change in the expression of the silenced genes , because they remain silenced . Increases in Pc and Rad21 binding at silenced genes upon cohesin and PRC1 depletion were substantially more frequent than decreases . Pc increases were detected at 60% of the H3K27me3-Pc genes upon Rad21 depletion , but decreases were detected at only 2% ( Figure 2C ) . At the H3K27me3-Pc genes that bind cohesin , some of which are not fully silenced , Pc increases were 2 . 5-fold more frequent than decreases upon cohesin depletion , which is opposite to what occurs at cohesin-binding active genes , where Pc decreases are 10-fold more frequent than increases ( Figure 2C ) . Thus Rad21 depletion causes coincident PRC1 decreases at active cohesin-binding genes and PRC1 increases at silenced genes , where cohesin binding is generally restricted to Polycomb Response Elements ( PREs; [12] ) . PRC1 binding may increase at silenced genes because it is released from active genes , increasing the amount available for binding . We cannot rule out the possibility , however , that cohesin directly competes with PRC1 for binding at genes marked by H3K27me3 , but that under normal conditions , the cohesin levels fall below the threshold for detection by ChIP . It is unlikely that cohesin is epitope-masked at silenced genes , because it was not detected outside of PREs at silenced genes with antibodies against Smc1 , SA , Rad21 and Nipped-B [12] . The antagonism between cohesin and PRC1 binding at PcG-silenced homeotic genes in the ANT-C and BX-C such as Scr and Antp ( Figure 3 ) provides a molecular explanation for dominant suppression of Pc mutant homeotic phenotypes by cohesin mutations [10] , [11] . The ectopic sex comb phenotype of heterozygous PRC1 mutants results from reduced silencing of Scr , and the abdominal segment transformations are caused by derepression of genes in the BX-C . We further confirmed the in vivo genetic antagonism by testing the effects of cohesin and releasin mutations on the phenotypes displayed by several PRC1 mutants . Smc1 and Rad21 cohesin mutations suppressed these phenotypes and pds5 releasin mutations enhanced them , supporting the idea that cohesin antagonizes PRC1 function at silenced genes during development ( Table S4 ) . The antagonism between cohesin and PRC1 binding at silenced genes predicts that heterozygous cohesin mutants would show phenotypic transformations opposite to those exhibited by PRC1 mutants because they would increase PcG silencing of homeotic genes . Indeed , we find that adult male flies heterozygous for Rad21 ( vtd ) loss-of-function mutations exhibit mild and weakly penetrant posterior to anterior abdominal transformations resulting in lighter pigmentation of abdominal segment A5 , which is an A5 to A4 transformation opposite to the A4 to A5 anterior to posterior transformation caused by PRC1 subunit mutations ( Figure S9 ) . The penetrance and extent of this transformation is increased by heterozygous Nipped-B kollerin mutations ( Figure S9 ) . This transformation indicates increased silencing of genes in BX-C , consistent with the increased PRC1 levels at these genes upon cohesin depletion in BG3 cells . Cohesin selectively binds active genes in which transcriptionally-engaged RNA polymerase II ( Pol II ) pauses several nucleotides downstream of the transcription start site [31] , [32] . One mechanism by which cohesin controls gene transcription is influencing the transition of paused Pol II to elongation . Cohesin depletion decreases the levels of total and elongating phosphorylated Pol II in the bodies of most cohesin-binding genes in BG3 cells , indicating that it often facilitates transition of paused Pol II to elongation [32] . Cohesin also hinders transition of paused Pol II to elongation at genes that are strongly repressed by cohesin , which include those rare genes such as inv and en in BG3 cells that have an extensive cohesin - H3K27me3 overlap [31] , [32] . Because cohesin facilitates PRC1 binding to active genes , we tested if PRC1 participates in the control of the transition of paused Pol II to elongation at active genes by mapping Pol II genome-wide before and after depletion of Ph in BG3 cells . We performed ChIP for Pol II subunit Rpb3 to measure total Pol II . Paused Pol II transitions to elongation after phosphorylation by P-TEFb [33] , and thus we also conducted ChIP for the Pol II Rpb1 subunit phosphorylated on the serine 2 residues of the heptapeptide repeats in the C terminal domain ( Ser2P Pol II ) to detect elongating Pol II . We correlated the effects of Ph depletion on Pol II occupancy with changes in mRNA levels measured by microarrays ( Table S3 ) . As described below , the results show that PRC1 influences transition of paused Pol II to elongation , but the effects of PRC1 depletion differ from those of cohesin depletion , indicating that cohesin has roles in controlling Pol II activity at active genes beyond facilitating PRC1 binding . They also confirm that PRC1 inhibits transcription of PcG-silenced genes . Plotting the change in Rpb3 levels versus the change in Ser2P Pol II levels shows that Ph depletion increased both Rpb3 and Ser2P Pol II at many PcG-silenced genes marked by H3K27me3 , consistent with the idea that PRC1 is essential for PcG-silencing ( Figure 3; Figure 4A; Figure S8 ) . The increase in Ser2P Pol II is often accompanied by an increase in mRNA ( Figure 4B , D ) . By contrast , at a large fraction of active genes , which lack H3K27me3 , Ph depletion increased total Pol II ( Rpb3 ) in the gene body , but decreased Ser2P Pol II ( Figure 3; Figure 4A; Figure S8 ) . The genes with increased total Pol II and decreased Ser2P Pol II often show decreased mRNA ( Figure 4B , C ) . Total Pol II increases and Ser2P Pol II decreases were much more frequent at cohesin-binding genes ( 40–50% ) than at genes that lack cohesin ( 12–15% ) , and at Pc-binding genes ( 30–40% ) than at genes that lack Pc ( 8–15% ) ( Figure S10 ) . Together , the findings that the total Pol II increases and Ser2P decreases upon Ph depletion ( a ) occur at a large fraction of active cohesin and Pc-binding genes , ( b ) happen more rarely at genes that don't bind cohesin , and ( c ) opposite changes ( Ser2P Pol II increases or Rpb3 decreases ) are rare , are compelling evidence that most of these Pol II changes are direct , and not caused indirectly by altered cell identity . Changes in cellular identity would be expected to affect a smaller fraction of active genes , and to alter Pol II occupancy more frequently in the opposite direction at cohesin and PRC1-binding genes and at non-binding genes . In summary , although total Rpb3 increased at both silent and active genes upon PRC1 depletion and rarely decreased , Ser2P Pol II and mRNA often increased at PcG-silenced genes and frequently decreased at active genes ( Figure 4A–D ) . The increase in total Pol II and decrease in Ser2P Pol II at many cohesin-binding active genes upon Ph PRC1 depletion argues that PRC1 blocks the release of non- or under-phosphorylated Pol II into elongation . Increases in total Pol II occurred 2 . 5-fold less frequently at promoters than in gene bodies upon Ph depletion , and the fold-increases in Pol II at promoters are also smaller ( Figure S10 ) . This argues that in most cases , transition to elongation , not Pol II recruitment or transcription initiation , is the key step regulated by PRC1 at active genes . Because promoter Pol II increases are more frequent than decreases , there may be modest effects on recruitment or initiation at some genes . The idea that PRC1 controls primarily the transition to elongation predicts that the pausing index , which is the ratio of total Pol II at the promoter to the total Pol II in the gene body will decrease upon Ph depletion . It also predicts that the ratio of phosphorylated Pol II to total Pol II in the gene body will decrease . Indeed , measuring the pause index and Ser2P to total Pol II ratio for all genes confirms global decreases in both cases , with a particularly dramatic decrease in the phosphorylated to total Pol II ratio in gene bodies ( Figure 4E , F ) . The strong decrease in the fraction of phosphorylated Pol II is consistent with the observed decreases in mRNA , because Pol II phosphorylation is required for association of elongation and RNA processing factors with elongating Pol II [34] . The under-phosphorylated Pol II that enters the gene bodies upon PRC1 depletion may have reduced processivity , and the nascent RNA is also unlikely to be efficiently spliced or polyadenylated . The effects of PRC1 depletion on Pol II activity at cohesin-binding genes differ substantially from those of cohesin depletion . Upon cohesion depletion , total Pol II and Ser2P Pol II levels usually change in the same direction , not in the opposite direction , with decreases being substantially more frequent than increases [32] . Thus the increase in total Pol II in the bodies of cohesin-binding genes upon PRC1 depletion likely requires cohesin , consistent with the finding that cohesin frequently facilitates the transition of paused Pol II to elongation [32] . Because the effects of cohesin depletion cannot be uncoupled from changes in transcription , we cannot rule out the possibility that the decreases in PRC1 upon cohesin depletion are caused in part by transcriptional changes instead of reduced recruitment or stabilization of PRC1 binding . We note , however , that the effects of cohesin reduction on PRC1 binding do not fully mirror the effects on Pol II occupancy , with the effects on PRC1 being more strongly skewed to the promoter . For instance , Rad21 depletion decreases Rpb3 at 12% of all active promoters [32] but decreases Pc on 18% of promoters , even though the starting Rpb3 signals are generally higher than Pc signals at promoters . In contrast , Rad21 depletion decreases Rpb3 in 50% of all active gene bodies , but Pc only in 35% . Thus reductions in Pol II levels at the promoter cannot explain all cases of reduced PRC1 . In anterior wing disc ( Figure 1 ) and cultured Sg4 cells [13] , the inv-en complex is PcG-silenced and has high H3K27me3 and PRC1 , but low cohesin . In the posterior wing disc , inv-en is expressed , with low H3K27me3 , but high cohesin and PRC1 ( Figure 1 ) . In BG3 cells , the inv-en complex has a different cohesin-PcG structure than either the silenced or active states [13] . H3K27me3 marks the entire complex , including the large flanking regulatory region , but inv and en and the regulatory region between them also bind cohesin ( Figure 5A ) . In these cells , inv and en mRNAs are present at modest levels and increase dramatically upon cohesin or PRC1 depletion [13] , [31] , [32] . Thus the inv-en complex in BG3 cells has a cohesin-PcG “restrained” state distinct from both the silenced and active states . Cohesin depletion can reduce long-range looping between enhancers and promoters , and between CTCF-binding sites in mammalian cells [4] , [35] . We thus considered the idea that part of the mechanism by which cohesin represses inv-en in BG3 cells is by facilitating looping between the Polycomb Response Elements ( PREs ) that recruit PcG complexes . There is a bipartite PRE just upstream of the en promoter , and another upstream of inv ( Figure 5; [36] ) . Other regulatory sequences include enhancers upstream of each gene , between the genes , and throughout the long region extending from en to the tou gene [37] . We used chromosome conformation capture ( 3C ) [38] to determine if the two PRE-containing regions in inv-en interact in BG3 cells , and if cohesin depletion reduces these interactions . We used four anchor restriction sites in a 180-kilobase region encompassing the inv and en genes , the regulatory regions , and flanking DNA ( Figure 5 ) . Two anchors ( a , d ) are outside the complex , and one ( c ) includes the en promoter and PREs . The fourth ( b ) is upstream of inv in the region containing a PRE . Figure 5A shows that the en PRE-promoter anchor ( c ) interacts with the inv PRE region . Similarly , the anchor near the inv PRE ( b ) interacts with the en PRE and with the region between inv and en , which harbors tissue-specific enhancers [37] . Both PRE anchors ( b , c ) also interact with the ends of the inv-en complex demarcated by the ends of the H3K27me3 domain . In contrast , the external anchors ( a , d ) interact only with their immediately surrounding regions , and not with sites in the complex . Upon depletion of the Rad21 cohesin subunit , which dramatically increased inv and en mRNA levels 20 to 40-fold [13] interaction between the inv and en PREs decreased substantially ( Figure 5A ) . Interaction of the inv PRE-containing region with the enhancer-containing region between inv and en also decreased , although the enhancer-containing region has lower , but significant cohesin levels . It may be that this interaction is stimulated secondarily by PRE-PRE looping . Facilitation of PRE-PRE looping by cohesin may explain why cohesin represses inv-en in BG3 cells . We cannot formally exclude , however , the possibility that cohesin represses by a different mechanism , and that the increased transcription caused by cohesin depletion reduces the PRE-PRE interaction . In Drosophila Sg4 cells in which inv-en is fully PcG-silenced , cohesin binds only at the two PREs , and there is no detectable Pol II or transcripts ( Figure 5B; [12] , [13] ) . We see interactions between the PREs as in BG3 cells , indicating that the PRE-PRE interactions seen in BG3 cells are consistent with PcG silencing ( Figure 5B ) . Interaction of the inv PRE with the end of the regulatory domain is lower than in BG3 cells . We were unable to test if cohesin depletion reduced the PRE-PRE interactions in the fully silenced state because Sg4 cells are relatively refractory to RNAi treatment [13] . However , cohesin also occupies all known PREs in the silenced BX-C in BG3 cells [12] , consistent with the possibility that cohesin can facilitate PRE-PRE looping at silenced genes .
PcG-silenced genes are targeted by both PRC1 and PRC2 , and generally do not bind measurable levels of cohesin except at PREs , but cohesin reduction increases Pc levels at PcG-silenced genes and PRC1 depletion increases cohesin binding . This antagonistic relationship contrasts with the functional interactions between PcG proteins and the Trithorax ( Trx ) protein , which binds PREs of silenced genes and is generally antagonistic to PcG silencing , but whose knockdown does not alter PRC1 binding [25] . Multiple mechanisms , which are not mutually exclusive , could contribute to the antagonism between PRC1 and cohesin association with PcG-silenced genes . Upon Ph PRC1 subunit depletion , increased transcription could facilitate cohesin binding . It might be expected , based on their direct interaction , that PRC1 would recruit cohesin to silenced genes , but PRC1 also has negative effects on cohesin binding at many active genes , and the presence of PRC2 at silenced genes may enhance these effects . For example , binding of Pc to H3K27me3 at silenced genes could alter PRC1 conformation such that the PRC1-cohesin contacts inhibit cohesin loading by kollerin , and/or facilitate cohesin removal by releasin . Cohesin depletion increases PRC1 levels at PcG-silenced genes , even though cohesin binding is very low at these genes outside of the PREs . One idea is that a reduction in cohesin dosage could release PRC1 from active genes , thereby making more PRC1 available to bind to silenced genes ( Figure 6 ) . Alternatively , cohesin and PRC1 binding could be dynamically competitive at silenced genes , with the competition favoring PRC1 , making it difficult to detect cohesin . Both models are consistent with the finding that a dominant wapl releasin mutation , which globally increases and stabilizes cohesin binding , causes similar mutant phenotypes as PRC1 mutations , indicating reduced PRC1 at Scr and other silenced genes [9] . Higher cohesin levels at active genes in the releasin mutant could sequester more PRC1 , making less available for binding to silenced genes . It more difficult to explain how more cohesin binding at silenced genes could directly reduce PRC1 binding in this mutant . One possibility is that binding of the Pc PRC1 subunit to H3K27me3 at silenced genes alters PRC1 conformation in a manner that allows cohesin to stimulate PRC1 removal instead of facilitate its binding . Because cohesin directly interacts with PRC1 and cohesin depletion reduces Pc binding at active genes , we posit that cohesin directly recruits and/or stabilizes PRC1 binding to active genes . As discussed below , our studies reveal that PRC1 controls Pol II phosphorylation and transition of paused Pol II to elongation at active genes . This may be similar to the function of PRC1 at some silenced and bivalent genes . Although the mechanisms by which PcG complexes repress transcription are not well understood , in some contexts , they inhibit transition of paused transcriptionally-engaged Pol II to active elongation [39]–[41] . Our findings suggest that at active genes , PRC1 facilitates phosphorylation of Pol II to the Ser2P Pol II elongating form and/or blocks entry of non-phosphorylated paused Pol II into elongation . Cohesin preferentially binds genes with promoter-proximal paused Pol II , and often facilitates , or less frequently , inhibits transition of paused Pol II to elongation [31] , [32] . Transition of paused Pol II to elongation requires phosphorylation of the NELF and DSIF pausing factors and Pol II by P-TEFb [33] . At most cohesin-binding genes , cohesin depletion reduces the levels of both total and phosphorylated Pol II in the gene body [32] . Here we find that PRC1 depletion , similar to cohesin depletion , reduces Ser2P Pol II at many active genes , but opposite to cohesin depletion , increases total Pol II in the same gene bodies . Thus both the global pausing index and ratio of phosphorylated Pol II to total Pol II in gene bodies decrease upon PRC1 depletion . This argues that Pol II enters into elongation with inadequate phosphorylation . One possible mechanism could be that PRC1 helps NELF and DSIF pausing factors restrain Pol II at the promoter until it is fully phosphorylated . The decrease in the ratio of phosphorylated to total Pol II upon PRC1 depletion is accompanied by a decrease in mRNA at many genes . This argues that the under-phosphorylated Pol II is less processive , and/or that the nascent RNA is not efficiently processed to mRNA . This is expected , because Pol II phosphorylation , in addition to facilitating transition to elongation , is required for binding of RNA processing factors to the transcription complex [34] . The fact that PRC1 depletion affects Pol II activity and mRNA levels at a large fraction of active cohesin-binding genes confirms that PRC1 is present and plays an important role at these genes . The finding that PRC1 directly controls transcription of many active genes is surprising because it is generally thought that PRC1 functions primarily at PcG-silenced genes . There is , however , published evidence consistent with our findings . Müller and colleagues used chromatin from mixed imaginal discs to perform Ph ChIP-chip with a different antibody than the one used here [28] , [42] . Although the mixture of many different cell types creates false overlaps , close examination of their data reveals that Ph associates with many constitutively active genes , including Act5C and some ribosomal protein genes , as we found in both wing discs and BG3 cells . Moreover , using different Pc , Psc and Ph antibodies , Paro and colleagues found that PRC1 preferentially binds promoters with paused RNA polymerase in cultured S2 cells [43] . Although the role of PRC1 in paused Pol II function was not investigated , this is consistent with the selective association of cohesin with genes that have paused Pol II [31] , [32] and our finding here that cohesin aids binding of PRC1 to active genes . There is also evidence that PcG complexes regulate some active genes in mammalian cells . In differentiated T helper cells , both PRC1 and PRC2 components are required for transcriptional activation of cytokine genes , although they repress HOX genes in the same cells [44] . Some expressed genes exhibit H3K27me3 , and the Ezh1 alternative methyltransferase is in PRC2 complexes that promote transcription during myogenic differentiation [45] , [46] . Ezh1 knockdown decreases Ser2P Pol II at the genes that it binds , similar to our findings with Ph depletion . We did not find any evidence , however , that PRC2 promotes transcription of active genes in Drosophila , which unlike mammals , has only one E ( z ) methyltransferase . The work presented here clearly establishes functional connections between cohesin and PRC1 , and suggests mechanisms for how cohesin and PRC1 control each other's activities and gene transcription . Our finding that three different cohesin-PcG states can exist at the invected-engrailed gene complex in different cells raises the question of how these various states arise . The presence or absence of PRC2 is likely to be one important factor that determines whether cohesin and PRC1 compete or collaborate , but the factors responsible for establishing these distinct states and controlling transitions between them remain to be determined .
Drosophila was cultured and genetic crosses were conducted at 25° as previously described [17] . The w1118; P{en2 . 4-GAL4}e16E , P{UAS-myr-mRFP}1 , P{NRE-EGFP . S}5A line [21] was used for preparing chromatin from anterior and posterior wing imaginal discs . Stocks were obtained from the Bloomington stock center . Nipped-B , pds5 , and cohesin mutant alleles have been described previously [10] , [17] , [47] . PcG mutant stocks were obtained from the Bloomington stock center and Rick Jones ( SMU ) . At least 30 male progeny from crosses between cohesin and PcG mutants were scored . Sex comb bristles on first , second , and third legs were counted , and abdominal transformations were scored using a defined arbitrary scale . Chromatin was prepared from late third instar Oregon-R wing imaginal discs according to Papp and Müller [48] , without dialysis . Sixty to seventy-five discs were used per immunoprecipitation , or 100 to 120 anterior or posterior segments . Chromatin preparation from BG3 cells , immunoprecipitation , and Affymetrix Drosophila 2 . 0R genome tiling microarray hybridization were as previously described [12] . Rick Jones ( SMU ) generously provided Pc antibodies ( Pc-RJ ) , and Vincenzo Pirrotta ( Rutgers ) kindly provided Pc ( Pc-VP ) , Ph , and Psc antibodies . Renato Paro ( ETH Zürich ) , Jürg Müller ( Max Planck Institute of Biochemistry ) , and Giacomo Cavalli ( Institute of Human Genetics ) provided additional Pc and Ph antibodies . Karen Adelman ( NIEHS ) generously provided Rpb3 antibodies . Ser2P Pol II and H3K27me3 antibodies were purchased from Abcam ( ab5095 , ab6002 ) and 8WG16 Pol II antibody was purchased from Covance ( MMS-126R ) . MAT software [19] was used to calculate ChIP enrichment . MAT scores measure enrichment relative to an input control and scale linearly with the log2 IP/control ratio . MAT has been experimentally demonstrated to be more sensitive and quantitative than other algorithms for measuring ChIP enrichment with Affymetrix tiling arrays , providing sensitivity equivalent to ChIP-seq at a density of 1 aligned read per genome base pair [18] , [49] . MAT uses probe sequence to perform within-array normalization , avoiding assumptions associated with quantile normalization . Bed files showing significant enrichment of proteins mapped by ChIP-chip ( Rad21 , Nipped-B , H3K27me3 , Pc-RJ , Pc-VP , Ph , Psc , Rpb3 , Ser2P Pol II ) at statistical thresholds of p≤10−3 or p≤10−2 were generated using MAT . Genes bound by a given protein were determined using the bed files and FlyBase gene annotations ( www . flybase . org , v5 . 28 ) with the Intersect tool of Galaxy/Cistrome [50] or BEDTools [51] . Venn diagrams were generated using eulerAPE ( www . eulerdiagrams . org/eulerAPE/ ) . Gene-based ChIP-chip signal data was extracted from MAT score files and aligned to gene expression data using custom programs . Data was analyzed using Microsoft Excel and R ( [52]; http://www . R-project . org ) . For some analyses , differences in integrated ChIP signals ( MAT scores ) over each annotated gene were calculated ( Figure S6 , method 1 ) . For others , increases and decreases in Rad21 , Pc , Rpb3 and Ser2P binding were measured by the difference in the ChIP MAT scores at each measured point in the genome between the RNAi treated samples and the controls ( Figure S6 , method 2 ) . Increases or decreases ≥2 standard deviations from the genome-wide median difference that extend for at least three contiguous microarray features ( 105 bp ) were used to generate bed files that were matched with gene annotations using BEDTools [51] . Examples of the increase ( + ) and decrease ( − ) bed files are shown in Figure 3 . Based on changes in Pol II occupancy upon cohesin depletion , these methods agree closely with results obtained by ChIP-qPCR at selected genes and genome-wide PRO-seq analysis [31] , [32] . Total RNA from wing discs or BG3 cells was isolated using Zymo ZR RNAi MicroPrep columns ( Zymo Research ) . Genome-wide analysis of wild-type wing disc mRNA with four biological replicates was performed using Affymetrix Drosophila GeneChip 2 . 0 microarrays as previously described [13] . Genome-wide measurements of three biological replicates of control BG3 cells and BG3 cells treated with Ph RNAi ( see below ) for five days was conducted in the same manner . BG3 cells were cultured and proteins were RNAi-depleted for Rad21 , Nipped-B or Ph as described [13] . The double-stranded RNAs used for Rad21 and Nipped-B RNAi were as described [13] . Two double-stranded RNAs targeting Polyhomeotic were used in concert; the constructs target regions of homology between polyhomeotic proximal and polyhomeotic distal . Primer sequences are as follows: Ph RNAi A , forward: TAATACGACTCACTATAGGGAGAAGCCATCAGCACCATGTCGC , reverse: TAATACGACTCACTATAGGGAGACGTAATTTCCGCCAGCGAATC . Ph RNAi B , forward: TAATACGACTCACTATAGGGAGATGCCCATTGATTCGCCCAAG , reverse: TAATACGACTCACTATAGGGAGATGCAACTTGTGGTAAAGGTGCC . These targets were designed using tools in the Drosophila RNAi Screening Center ( DRSC ) website ( www . flyrnai . org/ ) to avoid off-target effects . All RNAi-treated chromatin and 3C samples were collected 5 days after RNAi treatment . 3C was conducted using a modification of the strategy outlined by Miele et al . [53] . 108 BG3 or Sg4 cells were collected , washed with phosphate buffered saline , pH 7 . 6 ( PBS ) and then with hypotonic buffer [10 mM HEPES pH 7 . 9 , 50 mM NaCl , 1 mM dithiothreitol ( DTT ) , 10 mM MgCl2 , protease inhibitor cocktail ( cOmplete Mini , EDTA-Free Protease inhibitor tablets , Roche ) ] . Cells were lysed in hypotonic buffer containing 0 . 35 M sucrose and 0 . 2% NP-40 , vortexed for one minute , and immediately cross-linked with 1% formaldehyde for 10 minutes at room temperature . Cross-linking was stopped by adjustment to 135 mM glycine , and nuclei were isolated by centrifugation onto a 0 . 8 M sucrose cushion . Nuclei were washed in EcoRI restriction enzyme buffer , and 0 . 1% SDS was added to extract free protein . 1% Triton X-100 was used to sequester SDS , and EcoRI digestions ( 500–700 units ) were performed overnight at 37° . SDS was added to sequester EcoRI , and the digested DNA was ligated at 18° for 2 hours at a concentration of approximately 4 ng per microliter . Cross-linking was reversed by Proteinase K incubation at 65° for 6 hr or overnight , and DNA was isolated by sequential phenol , phenol-chloroform , and chloroform extractions , followed by precipitation with 0 . 5 volumes of 7 . 5 M ammonium acetate and 2 . 5 volumes of ethanol . DNA was dissolved in TE , digested with 500 ng per microliter RNAse A for 30 min at 37° and used for RT-PCR quantification . Digestion and religation of BAC DNA containing the entire inv-en locus was used as a normalization control ( BACR13F13; BACPAC Resources , Oakland , CA ) . This template was used to determine the amplification efficiency of each primer pair . EcoRI digestion efficiency was confirmed to be over 85% in several experiments using PCR primer pairs spanning the EcoRI sites . When the experiment was conducted without formaldehyde cross-linking , we were unable to detect 3C ligation products except between two immediate adjacent EcoRI sites ( likely resulting from incomplete digestion ) . Genome-wide ChIP and expression data generated for this study is deposited in the GEO database ( accession no . GSE42106 ) . | An important task for the cohesin protein complex that binds chromosomes is to ensure equal distribution of chromosomes into the daughter cells when a cell divides . Small changes in cohesin activity , however , can alter gene activity without affecting chromosome distribution , and disrupt physical and mental development . How cohesin controls gene activity and development is not well understood . In this study we show that cohesin controls the binding of the Polycomb Repressive Complex 1 ( PRC1 ) to many genes . PRC1 silences many genes that control development . Surprisingly , we find that cohesin aids binding of PRC1 to active genes , where PRC1 ensures that RNA polymerase , the enzyme that transcribes genes , is properly modified before entering the gene body . We also find that cohesin antagonizes the binding and activity of PRC1 at genes silenced by PRC1 , and can influence interactions between the DNA sequences that recruit PRC1 and other Polycomb complexes to silenced genes . These findings provide new and unexpected insights into how both cohesin and PRC1 control gene activity during development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"expression",
"analysis",
"animal",
"genetics",
"functional",
"genomics",
"gene",
"regulation",
"dna",
"transcription",
"histone",
"modification",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"molecular",
"genetics",
"epigenetics",
"epigenomics",
"chromatin",
"gene",
"expression",
"biology",
"microarrays",
"genetics",
"genomics",
"computational",
"biology"
] | 2013 | Cohesin and Polycomb Proteins Functionally Interact to Control Transcription at Silenced and Active Genes |
Timbre is the attribute of sound that allows humans and other animals to distinguish among different sound sources . Studies based on psychophysical judgments of musical timbre , ecological analyses of sound's physical characteristics as well as machine learning approaches have all suggested that timbre is a multifaceted attribute that invokes both spectral and temporal sound features . Here , we explored the neural underpinnings of musical timbre . We used a neuro-computational framework based on spectro-temporal receptive fields , recorded from over a thousand neurons in the mammalian primary auditory cortex as well as from simulated cortical neurons , augmented with a nonlinear classifier . The model was able to perform robust instrument classification irrespective of pitch and playing style , with an accuracy of 98 . 7% . Using the same front end , the model was also able to reproduce perceptual distance judgments between timbres as perceived by human listeners . The study demonstrates that joint spectro-temporal features , such as those observed in the mammalian primary auditory cortex , are critical to provide the rich-enough representation necessary to account for perceptual judgments of timbre by human listeners , as well as recognition of musical instruments .
A fundamental role of auditory perception is to infer the likely source of a sound; for instance to identify an animal in a dark forest , or to recognize a familiar voice on the phone . Timbre , often referred to as the color of sound , is believed to play a key role in this recognition process [1] . Though timbre is an intuitive concept , its formal definition is less so . The ANSI definition of timbre describes it as that attribute that allows us to distinguish between sounds having the same perceptual duration , loudness , and pitch , such as two different musical instruments playing exactly the same note [2] . In other words , it is neither duration , nor loudness , nor pitch; but is likely “everything else” . As has been often been pointed out , this definition by the negative does not state what are the perceptual dimensions underlying timbre perception . Spectrum is obviously a strong candidate: physical objects produce sounds with a spectral profile that reflects their particular sets of vibration modes and resonances [3] . Measures of spectral shape have thus been proposed as basic dimensions of timbre ( e . g . , formant position for voiced sounds in speech , sharpness , and brightness ) [4] , [5] . But timbre is not only spectrum , as changes of amplitude over time , the so-called temporal envelope , also have strong perceptual effects [6] , [7] . To identify the most salient timbre dimensions , statistical techniques such as multidimensional scaling have been used: perceptual differences between sound samples were collected and the underlying dimensionality of the timbre space inferred [8] , [9] . These studies suggest a combination of spectral and temporal dimensions to explain the perceptual distance judgments , but the precise nature of these dimensions varies across studies and sound sets [10] , [11] . Importantly , almost all timbre dimensions that have been proposed to date on the basis of psychophysical studies [12] are either purely spectral or purely temporal . The only spectro-temporal aspect of sound that has been considered in this context is related to the asynchrony of partials around the onset of a sound ( 8 , 9 ) , but the salience of this spectro-temporal dimension was found to be weak and context-dependent [13] . Technological approaches , not concerned with biology nor human perception , have explored much richer feature representations that span both spectral , temporal , and spectro-temporal dimensions . The motivation for these engineering techniques is an accurate recognition of specific sounds or acoustic events in a variety of applications ( e . g . automatic speech recognition; voice detection; music information retrieval; target tracking in multisensor networks and surveillance systems; medical diagnosis , etc . ) . Myriad spectral features have been proposed for audio content analysis , ranging from simple summary statistics of spectral shape ( e . g . spectral amplitude , peak , centroid , flatness ) to more elaborate descriptions of spectral information such as Mel-Frequency Cepstral Coefficients ( MFCC ) and Linear or Perceptual Predictive Coding ( LPC or PLP ) [14]–[16] . Such metrics have often been augmented with temporal information , which was found to improve the robustness of content identification [17] , [18] . Common modeling of temporal dynamics also ranged from simple summary statistics such as onsets , attack time , velocity , acceleration and higher-order moments to more sophisticated statistical temporal modeling using Hidden Markov Models , Artificial Neural Networks , Adaptive Resonance Theory models , Liquid State Machine systems and Self-Organizing Maps [19] , [20] . Overall , the choice of features was very dependent on the task at hand , the complexity of the dataset , and the desired performance level and robustness of the system . Complementing perceptual and technological approaches , brain-imaging techniques have been used to explore the neural underpinnings of timbre perception . Correlates of musical timbre dimensions suggested by multidimensional scaling studies have been observed using event-related potentials [21] . Other studies have attempted to identify the neural substrates of natural sound recognition , by looking for brain areas that would be selective to specific sound categories , such as voice-specific regions in secondary cortical areas [22] , [23] and other sound categories such as tools [24] or musical instruments [25] . A hierarchical model consistent with these findings has been proposed in which selectivity to different sound categories is refined as one climbs the processing chain [26] . An alternative , more distributed scheme has also been suggested [27] , [28] , which includes the contribution of low-level cues to the large perceptual differences between these high-level sound categories . A common issue for the psychophysical , technological , and neurophysiological investigations of timbre is that the generality of the results is mitigated by the particular characteristics of the sound set used . For multi-dimensional scaling behavioral studies , by construction , the dimensions found will be the most salient within the sound set; but they may not capture other dimensions which could nevertheless be crucial for the recognition of sounds outside the set . For engineering studies , dimensions may be designed arbitrarily as long as they afford good performance in a specific task . For the imaging studies , there is no suggestion yet as to which low-level acoustic features may be used to construct the various selectivity for high-level categories while preserving invariance within a category . Furthermore , there is a major gap between these studies and what is known from electrophysiological recordings in animal models . Decades of work have established that auditory cortical responses display rich and complex spectro-temporal receptive fields , even within primary areas [29] , [30] . This seems at odds with the limited set of spectral or temporal dimensions that are classically used to characterize timbre in perceptual studies . To bridge this gap , we investigate how cortical processing of spectro-temporal modulations can subserve both sound source recognition of musical instruments and perceptual timbre judgments . Specifically , cortical receptive fields and computational models derived from them are shown to be suited to classify a sound source from its evoked neural activity , across a wide range of instruments , pitches and playing styles , and also to predict accurately human judgments of timbre similarities
Responses in primary auditory cortex ( A1 ) exhibit rich selectivity that extends beyond the tonotopy observed in the auditory nerve . A1 neurons are not only tuned to the spectral energy at a given frequency , but also to the specifics of the local spectral shape such as its bandwidth [31] , spectral symmetry [32] , and temporal dynamics [33] ( Figure 1 ) . Put together , one can view the resulting representation of sound in A1 as a multidimensional mapping that spans at least three dimensions: ( 1 ) Best frequencies that span the entire auditory range; ( 2 ) Spectral shapes ( including bandwidth and symmetry ) that span a wide range from very broad ( 2–3 octaves ) to narrowly tuned ( <0 . 25 octaves ) ; and ( 3 ) Dynamics that range from very slow to relatively fast ( 1–30 Hz ) . This rich cortical mapping may reflect an elegant strategy for extracting acoustic cues that subserve the perception of various acoustic attributes ( pitch , loudness , location , and timbre ) as well as the recognition of complex sound objects , such as different musical instruments . This hypothesis was tested here by employing a database of spectro-temporal receptive fields ( STRFs ) recorded from 1110 single units in primary auditory cortex of 15 awake non-behaving ferrets . These receptive fields are linear descriptors of the selectivity of each cortical neuron to the spectral and temporal modulations evident in the cochlear “spectrogram-like” representation of complex acoustic signals that emerges in the auditory periphery . Such STRFs ( with a variety of nonlinear refinements ) have been shown to capture and predict well cortical responses to a variety of complex sounds like speech , music , and modulated noise [34]–[38] . To test the efficacy of STRFs in generating a representation of sound that can distinguish among a variety of complex categories , sounds from a large database of musical instruments were mapped onto cortical responses using the physiological STRFs described above . The time-frequency spectrogram for each note was convolved with each STRF in our neurophysiology database to yield a firing rate that is then integrated over time . This initial mapping was then reduced in dimensionality using singular value decomposition to a compact eigen-space; then augmented with a nonlinear statistical analysis using support vector machine ( SVM ) with Gaussian kernels [39] ( see METHODS for details ) . Briefly , support vector machines are classifiers that learn to separate , in our specific case , the patterns of cortical responses induced by the different instruments . The use of Gaussian kernels is a standard technique that allows to map the data from its original space ( where data may not be linearly separable ) onto a new representational space that is linearly separable . Ultimately , the analysis constructed a set of hyperplanes that outline the boundaries between different instruments . The identity of a new sample was then defined based on its configuration in this expanded space relative to the set of learned hyperplanes ( Figure 2 ) . Based on the configuration above and a 10% cross-validation technique , the model trained using the physiological cortical receptive fields achieved a classification accuracy of 87 . 22%±0 . 81 ( the number following the mean accuracy represents standard deviation , see Table 1 ) . Remarkably , this result was obtained with a large database of 11 instruments playing between 30 and 90 different pitches with 3 to 19 playing styles ( depending on the instrument ) , 3 style dynamics ( mezzo , forte and piano ) , and 3 manufacturers for each instrument ( an average of 1980 notes/instrument ) . This high classification accuracy was a strong indicator that neural processing at the level of primary auditory cortex could not only provide a basis for distinguishing between different instruments , but also had a robust invariant representation of instruments over a wide range of pitches and playing styles . Despite the encouraging results obtained using cortical receptive fields , the classification based on neurophysiological recordings was hampered by various shortcomings including recording noise and other experimental constraints . Also , the limited selection of receptive fields ( being from ferrets ) tended to under-represent parameter ranges relevant to humans such as lower frequencies , narrow bandwidths ( limited to a maximum resolution of 1 . 2 octaves ) , and coarse sampling of STRF dynamics . To circumvent these biases , we employed a model that mimics the basic transformations along the auditory pathway up to the level of A1 . Effectively , the model mapped the one-dimensional acoustic waveform onto a multidimensional feature space . Importantly , the model allowed us to sample the cortical space more uniformly than physiological data available to us , in line with findings in the literature [29] , [30] , [40] . The model operates by first mapping the acoustic signal into an auditory spectrogram . This initial transformation highlights the time varying spectral energies of different instruments which is at the core of most acoustic correlates and machine learning analyses of musical timbre [5] , [11] , [13] , [41] , [42] . For instance , temporal features in a musical note include fast dynamics that reflect the quality of the sound ( scratchy , whispered , or purely voiced ) , as well as slower modulations that carry nuances of musical timbre such as attack and decay times , subtle fluctuations of pitch ( vibrato ) or amplitude ( shimmer ) . Some of these characteristics can be readily seen in the auditory spectrograms , but many are only implicitly represented . For example , Figure 3A contrasts the auditory spectrogram of a piano vs . violin note . For violin , the temporal cross-section reflects the soft onset and sustained nature of bowing and typical vibrato fluctuations; the spectral slice captures the harmonic structure of the musical note with the overall envelope reflecting the resonances of the violin body . By contrast , the temporal and spectral modulations of a piano ( playing the same note ) are quite different . Temporally , the onset of piano rises and falls much faster , and its spectral envelope is much smoother . The cortical stage of the auditory model further analyzes the spectral and temporal modulations of the spectrogram along multiple spectral and temporal resolutions . The model projects the auditory spectrogram onto a 4-dimensional space , representing time , tonotopic frequency , spectral modulations ( or scales ) and temporal modulations ( or rates ) . The four dimensions of the cortical output can be interpreted in various ways . In one view , the cortical model output is a parallel repeated representation of the auditory spectrogram viewed at different resolutions . A different view is one of a bank of spectral and temporal modulation filters with different tuning ( from narrowband to broadband spectrally , and slow to fast modulations temporally ) . In such view , the cortical representation is a display of spectro-temporal modulations of each channel as they evolve over time . Ultimately each filter acts as a model cortical neuron whose output reflects the tuning of that neuronal site . The model employed here had 30 , 976 filters ( 128freq×22 rates×11 scales ) , hence allowing us to obtain a full uniform coverage of the cortical space and bypassing the limitations of neurophysiological data . Note that we are not suggesting that ∼30 K neurons are needed for timbre classification , as the feature space is reduced in further stages of the model ( see below ) . We have not performed an analysis of the number of neurons needed for such task . Nonetheless , a large and uniform sampling of the space seemed desirable . By collapsing the cortical display over frequency and averaging over time , one would obtain a two-dimensional display that preserves the “global” distribution of modulations over the remaining two dimensions of scale and rates . This “scale-rate” view is shown in Figure 3B for the same piano and violin notes in Figure 3A as well as others . Each instrument here is played at two distinct pitches with two different playing styles . The panels provide estimates of the overall distribution of spectro-temporal modulation of each sound . The left panel highlights the fact that the violin vibrato concentrates its peak energy near 6 Hz ( across all pitches and styles ) ; which matches the speed of pulsating pitch change caused by the rhythmic rate of 6 pulses per second chosen for the vibrato of this violin note . By contrast , the rapid onset of piano distributes its energy across a wider range of temporal modulations . Similarly , the unique pattern of peaks and valleys in spectral envelopes of each instrument produces a broad distribution along the spectral modulation axis , with the violin's sharper spectral peaks activating higher spectral modulations while the piano's smoother profile activates broad bandwidths . Each instrument , therefore , produces a correspondingly unique spectro-temporal activation pattern that could potentially be used to recognize it or distinguish it from others . Several computational models were compared in the same classification task analysis of the database of musical instruments as described earlier with real neurophysiological data . Results comparing all models are summarized in Table 1 . For what we refer to as the full model , we used the 4-D cortical model . The analysis started with a linear mapping through the model receptive fields , followed by dimensionality reduction and statistical classification using support vector machines with re-optimized Gaussian kernels ( see Methods ) . Tests used a 10% cross-validation method . The cortical model yielded an excellent classification accuracy of 98 . 7%±0 . 2 . We also explored the use of linear support vector machine , by bypassing the use of the Gaussian kernel . We performed a classification of instruments using the cortical responses obtained from the model receptive fields and a linear SVM . After optimization of the decision boundaries , we obtained an accuracy of 96 . 2%±0 . 5 . This result supports our initial assessment that the cortical space does indeed capture most of the subtleties that are unique to a common instrument but distinct between different classes . It is mostly the richness of the representation that underlies the classification performance: only a small improvement in accuracy is observed by adding the non-linear warping in the full model . In order to better understand the contribution of the cortical analysis beyond the time-frequency representation , we explored reduced versions of the full model . First we performed the timbre classification task using the auditory spectrogram as input . The feature spectra were obtained by processing the time waveform of each note through the cochlear-like filterbank front-end and averaging the auditory spectrograms over time , yielding a one-dimensional spectral profile for each note . These were then processed through the same statistical SVM model , with Gaussian functions optimized for this new representation using the exact same methods as used for cortical features . The classification accuracy for the spectral slices with SVM optimization attained a good but limited accuracy of 79 . 1%±0 . 7 . It is expected that a purely spectral model would not be able to classify all instruments . Whereas basic instrument classes differing by their physical characteristics ( wind , percussion , strings ) may have the potential to produce different spectral shapes , preserved in the spectral vector , more subtle differences in the temporal domain should prove difficult to recognize on this basis ( see Figure 4 ) . We shall revisit this issue of contribution and interactions between spectral and temporal features later ( see Control Experiments section ) . We performed a post-hoc analysis of the decision space based on cortical features in an attempt to get a better understanding of the configuration of the decision hyperplanes between different instrument classes . The analysis treated the support vectors ( i . e . samples of each instrument that fall right on the boundary that distinguishes it from another instrument ) for each instrument as samples from an underlying high-dimensional probability density function . Then , a measure of similarity between pairs of probability functions ( symmetric Kullback–Leibler ( KL ) divergence [43] ) was employed to provide a sense of distance between each instrument pair in the decision space . Because of the size and variability in the timbre decision space , we pooled the comparisons by instrument class ( winds , strings and percussions ) . We also focused our analysis on the reduced dimensions of the cortical space; called ‘eigen’-rate , ‘eigen’-scale and ‘eigen’-frequencies; obtained by projecting the equivalent dimensions in the cortical tensor ( rate , scale and frequency , respectively ) into a reduced dimensional space using singular-value decomposition ( see METHODS ) . The analysis revealed a number of observations ( see Figure 5 ) . For instance , wind and percussion classes were the most different ( occupy distant regions in the decision space ) , followed by strings and percussions then strings and winds ( average KL distances were 0 . 58 , 0 . 41 , 0 . 35 , respectively ) . This observation was consistent with the subjective judgments of human listeners presented next ( see off-diagonal entries in Figure 6B ) . All 3 pair comparisons were statistically significantly different from each other ( Wilcoxon ranksum test , p<10−5 for all 3 pairs ) . Secondly , the analysis revealed that the 2 first ‘eigen’-rates captured most of the difference between the instrument classes ( statistical significance in comparing the first 2 eigenrates with the others; Wilcoxon ranksum test , p = 0 . 0046 ) . In contrast , all ‘eigen’-scales were variable across classes ( Kruskal-Wallis test , p = 0 . 9185 indicating that all ‘eigen’-scales contributed equally in distinguishing the broad classes ) . A similar analysis indicated that the first four ‘eigen’-frequencies were also significantly different from the remaining features ( Wilcoxon ranksum test , p<10−5 ) . One way to interpret these observations is that the first two principal orientations along the rate axis captured most of the differences that distinguish winds , strings and percussions . This seems consistent with the large differences in temporal envelope shape for these instruments classes , which can be represented by a few rates . By contrast , the scale dimension ( which captures mostly spectral shape , symmetry and bandwidth ) was required in its entirety to draw a boundary between these classes , suggesting that unlike the coarser temporal characteristics , differentiating among instruments entails detailed spectral distinctions of a subtle nature . Spectral features have been extensively used for tasks of musical timbre classification of isolated notes , solo performances or even multi-instrument recordings . Features such as Cepstral Coefficients or Linear Prediction of the spectrum resonances yielded performance in the range of 77% to 90% when applied to databases similar to the one used in the present study [44]–[46] . There is wide agreement in the literature that inclusion of simple temporal features , such as zero-crossing rate , or more complex ones such as trajectory estimation of spectral envelopes , is often desirable and results in improvement of the system performance . Tests on the RWC database with both spectral and temporal features reported an accuracy of 79 . 7% using 19 instruments [47] or 94 . 9% using 5 instruments [42] . Tests of spectrotemporal features on other music databases has often yielded a range of performances between 70–95% [48]–[51] . Whereas a detailed comparisons with our results is beyond the scope of this paper , we can still note that , if anything , the recognition rates we report for the full auditory model are generally in the range or above those reported by state-of-the-art signal processing techniques . Given the ability of the cortical model to capture the diversity of musical timbre across a wide range of instruments in a classification task , we next explored how well the cortical representation ( from both real and model neurons ) does in capturing human perceptual judgments of distance in the musical timbre space . To this end , we used human judgments of musical timbre distances using a psychoacoustic comparison paradigm . Human listeners were asked to rate the similarity between musical instruments . We used three different notes ( A3 , D4 and G#4 ) in three different experiments . Similarity matrices for all three notes yielded reasonably balanced average ratings across subjects , instrument pair order ( e . g . piano/violin vs . violin/piano ) and pitches , in agreement with other studies [52] ( Figure 6A ) . Therefore , we combined the matrices across notes and listeners into an upper half matrix shown in Figure 6B , and used it for all subsequent analyses . For comparison with previous studies , we also ran a multidimensional scaling ( MDS ) analysis [53] on this average timbre similarity rating and confirmed that the general configuration of the perceptual space was consistent with previous studies ( Figure 6C ) [8] . Also for comparison , we tested acoustical dimensions suggested in those studies . The first dimension of our space correlated strongly with the logarithm of attack-time ( Pearson's correlation coefficient: ρ = 0 . 97 , p<10−3 ) , and the second dimension correlated reasonably well with the center of mass of the auditory spectrogram , also known as spectral centroid ( Pearson's correlation coefficient: ρ = 0 . 62 , p = 0 . 04 ) . The perceptual results obtained above , reflecting subjective timbre distances between different instruments , summarizes an elaborate set of judgments that potentially reveal other facets of timbre perception than the listeners' ability to recognize instruments . We then explored whether the cortical representation could account for these judgments . Specifically , we asked whether the cortical analysis maps musical notes onto a feature space where instruments like violin and cello are distinct , yet closer to each other than a violin and a trumpet . We used the same 11 instruments and 3 pitches ( A3 , D4 and G#4 ) employed in the psychoacoustics experiment above and mapped them onto a cortical representation using both neurophysiological and model STRFs . Each note was then vectorized into a feature data-point and mapped via Gaussian kernels . These kernels are similar to the radial basis functions used in the previous section , and aimed at mapping the data from its original cortical space to a linearly separable space . Unlike the generic SVM used in the classification of musical timbre , the kernel parameters here were optimized based on the human scores following a similarity-based objective function . The task here was not merely to classify instruments into distinct classes , but rather to map the cortical features according to a complex set of rules . Using this learnt mapping , a confusion matrix was constructed based on the instrument distances , which was then compared with the human confusion matrix using a Pearson's correlation metric . We performed a comparison with the physiological as well as model STRFs . The simulated confusion matrices are shown in Figure 7A–B . The success or otherwise of the different models was estimated by correlating the human dissimilarity matrix to that generated by the model . No attempt was made at producing MDS analyses of the model output , as meaningfully comparing MDS spaces is not a trivial problem [52] . Physiological STRFs yielded a correlation coefficient of 0 . 73 , while model STRFs yielded a correlation of 0 . 94 ( Table 2 ) . In order to disentangle the contribution of the “input” cortical features versus the “back-end” machine learning in capturing human behavioral data , we recomputed confusion matrices using alternative representations such as the auditory spectrogram and various marginals of the cortical distributions . In all these control experiments , the Gaussian kernels were re-optimized separately to fit the data representation being explored . We first investigated the performance using auditory spectrum features with optimized Gaussian kernels . The spectrogram representation yielded a similarity matrix that captures the main trends in human distance judgments , with a correlation coefficient of 0 . 74 ( Figure 7C , leftmost panel ) . Similar experiments using a traditional spectrum ( based on Fourier analysis of the signal ) yield a correlation of 0 . 69 . Next , we examined the effectiveness of the model cortical features by reducing them to various marginal versions with fewer dimensions as follows . First , we performed an analysis of the spectral and temporal modulations as a separable cascade of two operations . Specifically , we analyzed the spectral profile of the auditory spectrogram ( scales ) independently from the temporal dynamics ( rates ) and stack the two resulting feature vectors together . This analysis differed from the full cortical analysis that assumes an inseparable analysis of spectro-temporal features . An inseparable function is one that cannot be factorized into a function of time and a function of frequency; i . e . a matrix of rank greater than 1 ( see Methods ) . By construction , a separable function consists of temporal cross sections that are scaled versions of the same essential temporal function . A consequence of such constraint is that a separable function cannot capture orientation in time-frequency space ( e . g . FM sweeps ) . In contrast , the full cortical analysis estimates modulations along both time and frequency axes in addition to an integrated view of the two axes including orientation information The analysis based on the separable model achieved a correlation coefficient of 0 . 83 ( Table 2 ) . Second , we further reduced the separable spectro-temporal space by analyzing the modulation content along both time and frequency without maintaining the distribution along the tonotopic axis . This was achieved by simply integrating the modulation features along the spectral axis thus exploring the global characteristic of modulation regardless of tonotopy ( Figure 7C , rightmost panel ) . This representation is somewhat akin to what would result from a 2-dimensional Fourier analysis of the auditory spectrogram . This experiment yielded a correlation coefficient of 0 . 70 ( Table 2 ) , supporting the value of an explicit tonotopic axis in capturing subtle difference between instruments . Next , we addressed the concern that the mere number of features included in the full cortical model enough to explain the observed performance . We therefore undersampled the full cortical model by employing only 6 scale filters; 10 rate filters and 64 frequency filters by coarsely sampling the range of spectro-temporal modulations . This mapping resulted in a total number of dimensions of 3840; to be comparable to the 4224 dimensions obtained from the separable model . We then performed the dimensionality reduction to 420 dimensions , similar to that used for the separable analysis discussed above . The correlation obtained was 0 . 86; which is better than that of the separable spectro-temporal model ( see Figure 8 ) . This result supports our main claim that the coverage provided by the cortical space allows extracting specific details in the musical notes that highlight information about the physical properties of each instrument; hence enabling classification and recognition . Finally , we examined the value of the kernel-learning compared to using a simple Euclidian L2 distance at various stages of the model ( e . g . peripheral model , cortical stage , reduced cortical model using tensor singular value decomposition ) . Table 2 summarizes the results of this analysis along various stages of the model shown in Figure 2 . The analysis revealed that the kernel-based mapping does provide noticeable improvement to the predictive power of the model but cannot –by itself– explain the results since the same technique applied directly on the spectrum only yielded a correlation of 0 . 74 .
This study demonstrates that perception of musical timbre could be effectively based on neural activations patterns that sounds evoke at the level of primary auditory cortex . Using neurophysiological recordings in mammalian auditory cortex as well as a simplified model of cortical processing , it is possible to accurately replicate human perceptual similarity judgments and classification performance among sounds from a large number of musical instruments . Of course , showing that the information is available at the level of primary auditory cortex does not imply that all neural correlates of sound identification will be found at this level . Nevertheless , it suggests that the spectro-temporal transforms as observed at this stage are critical for timbre perception . Moreover , our analysis highlights the ability of the cortical mapping to capture timbre properties of musical notes and instrument-specific characteristics regardless of pitch and playing style . Unlike static or reduced views of timbre that emphasize three or four parameters extracted from the acoustic waveform , the cortical analysis provides a dynamic view of the spectro-temporal modulations in the signal as they vary over time . A close examination of the contribution of different auditory features and processing stages to the timbre percepts highlights three key points . First , neither the traditional spectrum nor its variants ( e . g . average auditory spectrum [54] ) are well-suited to account for timbre perception in full . According to our simulations , these representations encode the relevant spectral and temporal acoustic features too implicitly to lend themselves for exploitation by classifiers and other machine learning techniques . In some sense , this conclusion is expected given the multidimensional nature of the timbre percept compared to the dense two-dimensional spectrogram; and is in agreement with other findings from the literature [19] . Second , when considering more elaborate spectro-temporal cortical representations , it appears that the full representation accounts best for human performance . The match worsens if instead marginals are used by collapsing the cortical representation onto one or more dimensions to extract the purely spectral or temporal axes or scale-rate map ( Figure 3 , Tables 1 and 2 ) . This is the case even if all dimensions are used separately , suggesting that there are joint spectro-temporal features that are key to a full accounting of timbre . While the role of both purely spectral and temporal cues in musical timbre is quite established [12] , our analysis emphasizes the crucial contribution of a joint spectro-temporal representation . For example , FM modulations typical of vibrato in string instruments are joint features that cannot be easily captured by the marginal spectral or temporal representations . Interestingly , acoustical analyses and fMRI data in monkeys suggest that the spectro-temporal processing scheme used here may be able to differentiate between broad sound categories ( such as monkey calls vs . bird calls vs . human voice ) , with corresponding neural correlates when listening to those sounds [55] . Third , a nonlinear decision boundary in the SVM classifier is essential to attain the highest possible match between the cortical representation and human perception . Linear metrics such as L2 are less optimal , indicating that the linear cortical representation may not be sufficiently versatile to capture the nuances of various timbres . The inadequacy of the linear cortical mapping has previously been described when analyzing neural responses to complex sounds such as speech at the level of auditory cortex [35] , [36] , [38] . In these cases , it is necessary to postulate the existence of nonlinearities such as divisive normalization or synaptic depression that follows a linear spectro-temporal analysis so as to account fully for the observed responses . In the current study , the exact nature of the nonlinearity remains unclear as it is implicitly subsumed in the Gaussian kernels and subsequent decisions . In summary , this study leads to the general conclusion that timbre percepts can be effectively explained by the joint spectro-temporal analysis performed at the level of mammalian auditory cortex . However , unlike the small number of spectral or temporal dimensions that have been traditionally considered in the timbre literature , we cannot highlight a simple set of neural dimensions subserving timbre perception . Instead , the model suggests that subtle perceptual distinctions exhibited by human listeners are based on ‘opportunistic’ acoustic dimensions [56] that are selected and enhanced , when required , on the rich baseline provided by the cortical spectro-temporal representation .
All behavioral recordings of timbre similarity judgments with human listeners were approved by the local ethics committee of the Université Paris Descartes . All procedures for recordings of single unit neural activity in ferrets were in accordance with the Institutional Animal Care and Use Committee at the University of Maryland , College Park and the Guidelines of the National Institutes of Health for use of animals in biomedical research . The cortical model is comprised of two main stages: an early stage mimicking peripheral processing up to the level of the midbrain , and a central stage capturing processing in primary auditory cortex ( A1 ) . Full details about the model can be found in [54] , [58]; but are described briefly here . The processing of the acoustic signal in the cochlea is modeled as a bank of 128 constant-Q asymmetric bandpass filters equally spaced on the logarithmic frequency scale spanning 5 . 3 octaves . The cochlear output is then transduced into inner hair cells potentials via a high pass and low pass operation . The resulting auditory nerve signals undergo further spectral sharpening via a lateral inhibitory network . Finally , a midbrain model resulting in additional loss in phase locking is performed using short term integration with time constant 4 ms resulting in a time frequency representation called as the auditory spectrogram . The central stage further analyzes the spectro-temporal content of the auditory spectrogram using a bank of modulation selective filters centered at each frequency along the tonotopic axis , modeling neurophysiological receptive fields . This step corresponds to a 2D affine wavelet transform , with a spectro-temporal mother wavelet , define as Gabor-shaped in frequency and exponential in time . Each filter is tuned ( Q = 1 ) to a specific rate ( in Hz ) of temporal modulations and a specific scale of spectral modulations ( in cycles/octave ) , and a directional orientation ( + for upward and − for downward ) . For input spectrogram the response of each STRF in the model is given by: ( 1 ) where denotes convolution in time and frequency and and are the characteristic phases of the STRF's which determine the degree of asymmetry in the time and frequency axes respectively . The model filters filters can be decomposed in each quadrant ( upward + or downward − ) into into corresponding to rate and scale filters respectively . Details of the design of the filter functions can be found in [58] . The present study uses 11 spectral filters with characteristic scales [0 . 25 , 0 . 35 , 0 . 50 , 0 . 71 , 1 . 00 , 1 . 41 , 2 . 00 , 2 . 83 , 4 . 00 , 5 . 66 , 8 . 00] ( cycles/octave ) and 11 temporal filters with characteristic rates [4 . 0 , 5 . 7 , 8 . 0 , 11 . 3 , 16 . 0 , 22 . 6 , 32 . 0 , 45 . 3 , 64 . 0 , 90 . 5 , 128 . 0] ( Hz ) , each with upward and downward directionality . All outputs are integrated over the time duration of each note . In order to simplify the analysis , we limit our computations to the magnitude of the cortical output ( i . e . responses corresponding to zero-phase filters ) . Finally , dimensionality reduction is performed using tensor singular-value decomposition [59] . This technique unfolds the cortical tensor along each dimension ( frequency , rate and scale axes ) and applies singular value decomposition on the unfolded matrix . We choose 5 eigenscales , 4 eigenrates and 21 eignefrequencies resulting in 420 features with the highest eigenvalues , preserving 99 . 9% of the variance in the original data . The motivation for this cutoff choice is presented later . Data used here was collected in the context of a number of studies [60]–[62] and full details of the experimental paradigm are described in these publications . Briefly , extracellular recordings were performed in 15 awake non-behaving domestic ferrets ( Mustela putorius ) with surgically implanted headposts . Tungsten electrodes ( 3–8 MΩ ) were used to record neural responses from single and multi-units at different depths . All data was processed off-line and sorted to extract single-unit activity . Spectro-Temporal Receptive fields ( STRF ) were characterized using TORC ( Temporally-Orthogonal Ripple Combination ) stimuli [63] , consisting of superimposed ripple noises with rates between 4–24 ( Hz ) and scales between 0 ( flat ) and 1 . 4 peaks/octave . Each stimulus was 3 sec with inter-stimulus intervals of 1–1 . 2 sec , and a full set of 30 TORCs was typically repeated 6–15 times . All sounds were computer-generated and delivered to the animal's ear through inserted earphones calibrated in-situ . TORC amplitude is fixed between 55–75 dB SPL . STRFs were derived using standard reverse correlation techniques , and a signal-to-noise ratio ( SNR ) for each STRF was measured using a bootstrap technique ( see [63] for details ) . Only STRFs with SNR≥2 were included in the current study , resulting in a database of 1110 STRFs ( average 74 STRFs/animal ) . Note because of the experimental paradigm , STRFs spanned a 5-octave range with low frequencies 125 , 250 or 500 Hz . In the current study , all STRFs were aligned to match the frequency range of musical note spectrograms . Since all our spectrograms start at 180 Hz and cover 5 . 3 octaves , we scaled and shifted the STRF's to fit this range . The neurophysiological STRFs were employed to perform the timbre analysis by convolving each note's auditory spectrogram z ( t , f ) with each STRF in the database as in Equation ( 2 ) . ( 2 ) The resulting firing rate vector was then integrated over time yielding an average response across the tonotopic axis . The output from all STRFs were then stacked together , resulting in a 142080 ( 128 frequency channels ×1110 STRFs ) dimensional vector . We reduced this vector using singular value decomposition and mapped it onto 420 dimensions , which preserve 99 . 9% of the data variance in agreement with dimensionality used for model STRFs . In order to test the cortical representation's ability to discriminate between different musical instruments , we augmented the basic auditory model with a statistical clustering model based on support vector machines ( SVM ) [39] . Support vector machines are classifiers that learn a set of hyperplanes ( or decision boundaries ) in order to maximally separate the patterns of cortical responses caused by the different instruments . Each cortical pattern was projected via Gaussian kernel to a new dimensional space . The use of kernels is a standard technique used with support vector machines , aiming to map the data from its original space ( where data may not be linearly separable ) onto a new representational space that is linearly separable . This mapping of data to a new ( more linear space ) through a the use of a kernel or transform is commonly referred to as the “kernel trick” [39] . In essence , kernel functions aim to determine the relative position or similarity between pairs of points in the data . Because the data may lie in a space that is not linearly separable ( not possible to use simple lines or planes to separate the different classes ) , it is desirable to map the data points onto a different space where this linear separability is possible . However , instead of simply projecting the data points themselves onto a high-dimensional feature space which would increase complexity as a function of dimensionality , the “kernel trick” avoids this direct mapping . Instead , it provides a method for mapping the data into an inner product space without explicitly computing the mapping of the observations directly . In other words , it computes the inner product between the data points in the new space without computing the mapping explicitly . The kernel used here is given by ( 3 ) where and are the feature vectors of 2 sound samples . The parameter for the Gaussian kernel and the cost parameter for the SVM algorithm were optimized on a subset of the training data . A classifier is trained for every pair of classes and . Each of these classifiers then gives a label for a test sample . Note that or . We count the number of labels . The test sample is then assigned to the class with maximum count given by . The parameter in Equation ( 3 ) was chosen by doing a grid search over a large parameter span in order to optimize the classifier performance in correctly distinguishing different instruments . This tuning was done by training and testing on a subset of the training data . For model testing , we performed a standard k-fold cross validation procedure with k = 10 ( 90% training , 10% testing ) . The dataset was divided into 10 parts . We then left out one part at a time and trained on the remaining 9 parts . The results reported are the average performance over all 10 iterations . A single Gaussian parameter was optimized for all the pair-wise classifiers across all the 10-fold cross validation experiments . In order to better understand the mapping of the different notes in the high-dimensional space used to classify them , we performed a closer analysis of the support vectors for each instrument pair i and j . Support vectors are the samples from each class that fall exactly on the margin between class i and class j , and therefore are likely to be more confusable between the classes . Since we are operating in the ‘classifier space’ , each of the support vectors is defined in a reduced dimensional hyperspace consisting of 5 eigen-scales , 4 eigen-rates , and 21 eigen-frequencies as explained above ( a total of 420 dimensions ) . The collection of all support vectors for each class i can be pulled together to estimate a high-dimensional probability density function . The density function estimate was derived using a histogram method by partitioning the sample space along each dimension into 100 bins , counting how many samples fall into each bin and dividing the counts by the total number of samples . We label the probability distribution for the d-th dimension ( d = 1 , . . , 420 ) . We then computed the symmetric KL divergence , [43] , between the support vectors for classes and from the classifier as shown in Equation ( 4 ) . The KL divergence is simply a measure of difference between pairs of probability distributions , is defined is next: ( 4 ) The bins with zero probability were disregarded from the computation of the KL divergence . An alternative method that smoothed the probability distribution over the zero bins was also tested and yielded virtually comparable results . Overall , this analysis is meant to inform about the layout of the timbre decision space . We analyzed the significance of the results between the broad timbre classes ( winds , percussions and strings ) by pooling individual comparisons between instruments within each group ( See Figure 5 ) . We tested the auditory model's ability to predict human listeners' judgment of musical timbre distances . Just like the timbre classification task , we used the cortical model augmented with Gaussian Kernels . In order to optimize the model to the test data , we employed a variation of the Gaussian kernel that performs an optimized feature embedding on every data dimension . The kernel is defined as follows: ( 5 ) where N is the number of dimensions of the features x and y . 's are parameters for the kernel that need to be optimized . We define an objective function that optimizes the correlation between the human perceptual distances and the distances in the embedded space . ( 6 ) where is the average profile for the ith instrument over all notes; D ( i , j ) is the average perceived distance between the ith and jth instrument based on psychoacoustic results and are the average distances from the kernel and the psychoacoustic experiment respectively . represents the variance of the kernel distances over all samples ( all instrument pairs ) . Similarly is the variance of the human perceived distances . We used a gradient ascent algorithm to learn which optimize the objective function . The correlation analysis employed the same dataset used for the human psychophysical experiment described above . Each note was 0 . 25 s in duration with sampling rate 44 . 1 kHz and underwent the same preprocessing as mentioned earlier . The absolute value of the model output was derived for each note and averaged over duration following a similar procedure as the timbre classification described above . The cortical features obtained for the three notes ( A3 , D4 , G#4 ) were averaged for each instrument i to obtain . Similarly the perceived human distances between instrument i and j were obtained by averaging the ( i , j ) th and ( j , i ) th entry in the human distance matrix over all the 3 notes to obtain D ( i , j ) . Finally , the human and model similarity matrices were compared using the Pearson's correlation metric . In order to avoid overestimating the correlation between the two matrices ( the two symmetric values appearing twice in the correlation ) , we correlated only the upper triangle of each matrix . | Music is a complex acoustic experience that we often take for granted . Whether sitting at a symphony hall or enjoying a melody over earphones , we have no difficulty identifying the instruments playing , following various beats , or simply distinguishing a flute from an oboe . Our brains rely on a number of sound attributes to analyze the music in our ears . These attributes can be straightforward like loudness or quite complex like the identity of the instrument . A major contributor to our ability to recognize instruments is what is formally called ‘timbre’ . Of all perceptual attributes of music , timbre remains the most mysterious and least amenable to a simple mathematical abstraction . In this work , we examine the neural underpinnings of musical timbre in an attempt to both define its perceptual space and explore the processes underlying timbre-based recognition . We propose a scheme based on responses observed at the level of mammalian primary auditory cortex and show that it can accurately predict sound source recognition and perceptual timbre judgments by human listeners . The analyses presented here strongly suggest that rich representations such as those observed in auditory cortex are critical in mediating timbre percepts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"auditory",
"system",
"computational",
"neuroscience",
"psychoacoustics",
"biology",
"sensory",
"systems",
"sensory",
"perception",
"neuroscience"
] | 2012 | Music in Our Ears: The Biological Bases of Musical Timbre Perception |
Rickettsial agents are sensed by pattern recognition receptors but lack pathogen-associated molecular patterns commonly observed in facultative intracellular bacteria . Due to these molecular features , the order Rickettsiales can be used to uncover broader principles of bacterial immunity . Here , we used the bacterium Anaplasma phagocytophilum , the agent of human granulocytic anaplasmosis , to reveal a novel microbial surveillance system . Mechanistically , we discovered that upon A . phagocytophilum infection , cytosolic phospholipase A2 cleaves arachidonic acid from phospholipids , which is converted to the eicosanoid prostaglandin E2 ( PGE2 ) via cyclooxygenase 2 ( COX2 ) and the membrane associated prostaglandin E synthase-1 ( mPGES-1 ) . PGE2-EP3 receptor signaling leads to activation of the NLRC4 inflammasome and secretion of interleukin ( IL ) -1β and IL-18 . Importantly , the receptor-interacting serine/threonine-protein kinase 2 ( RIPK2 ) was identified as a major regulator of the immune response against A . phagocytophilum . Accordingly , mice lacking COX2 were more susceptible to A . phagocytophilum , had a defect in IL-18 secretion and exhibited splenomegaly and damage to the splenic architecture . Remarkably , Salmonella-induced NLRC4 inflammasome activation was not affected by either chemical inhibition or genetic ablation of genes associated with PGE2 biosynthesis and signaling . This divergence in immune circuitry was due to reduced levels of the PGE2-EP3 receptor during Salmonella infection when compared to A . phagocytophilum . Collectively , we reveal the existence of a functionally distinct NLRC4 inflammasome illustrated by the rickettsial agent A . phagocytophilum .
Rickettsial diseases are arthropod-borne illnesses caused by obligate intracellular bacteria grouped in the order Rickettsiales [1 , 2] . They include: ( i ) rickettsioses due to bacteria of the genus Rickettsia , including the spotted fever and the typhus group; ( ii ) scrub typhus due to Orientia tsutsugamushi; and ( iii ) ehrlichioses and anaplasmosis due to bacteria within the family Anaplasmataceae [1 , 2] . Some aspects of rickettsial recognition by the immune system have been described [1 , 2] . For instance , Rickettsia spp . have a structurally distinct form of lipopolysaccharide ( LPS ) that appears identifiable by Toll-like receptor ( TLR ) 4 [2–5] , whereas the TLR2-MyD88 ( Myeloid Differentiation Primary Response Protein 88 ) axis plays a critical role in host defense against ehrlichial infection [6 , 7] . However , how these organisms are sensed by pattern recognition receptors ( PRRs ) remains mostly undefined . Bona fide pathogen-associated molecular patterns ( PAMPs ) are conspicuously absent in some of these microbes when compared to classically-defined bacterial pathogens [2 , 8–10] . As an example , Anaplasma and Ehrlichia spp . are considered Gram-negative bacteria , but are unable to synthesize LPS or peptidoglycans [8 , 9 , 11] . Additionally , O . tsutsugamushi does not carry genes in its genome for producing lipid A and has no LPS [10 , 12] . Counterintuitively , three independent groups have demonstrated that the NOD ( Nucleotide-Binding Oligomerization Domain Protein ) -RIPK2 ( Receptor-Interacting Serine/Threonine-Protein Kinase 2 ) pathway , which recognizes peptidoglycans [13] , were important to combat Ehrlichia , Anaplasma and Orientia spp . infection [6 , 14 , 15] . Furthermore , the non-canonical caspase-11 inflammasome , the molecular scaffold that senses LPS in the cytosol and regulates inflammatory cell death or pyroptosis [16] , was shown to mediate Ehrlichia-induced immunopathology [17] . Nonetheless , Ehrlichia spp . do not carry genes for the biosynthesis of LPS in their genomes [11] , and are neither cytosolic bacteria nor do they trigger pyroptosis [8] . Mice deficient in NLRC4 [NOD-like receptor ( NLR ) containing a caspase activating and recruitment domain ( CARD ) 4] , the adaptor molecule that is engaged by NAIP ( Neuronal apoptosis inhibitory protein ) receptors upon recognition of the bacterial type III secretion system ( T3SS ) and flagellin [18–24] , are also susceptible to A . phagocytophilum [25] . Importantly , A . phagocytophilum is aflagellated and does not have a T3SS [9 , 26] . These findings suggest that the life style of rickettsial agents induces a mode of immune recognition , which can be exploited for the discovery of unique pathogen-sensing systems . Previously , we discovered that mice deficient in Nlrc4 and Caspase-1/11 are susceptible to A . phagocytophilum infection [25] . We also reported that A . phagocytophilum causes NLRC4 inflammasome activation and caspase-1 autoproteolysis through the phospholipid-binding protein Annexin A2 [27 , 28] . The mechanistic delineation of how the NLRC4 inflammasome was induced remained elusive . In this article , we show a novel mode of NLRC4 inflammasome circuitry that is dependent on the eicosanoid prostaglandin E2 ( PGE2 ) . Upon A . phagocytophilum infection , cytosolic phospholipase A2 ( cPLA2 ) cleaves arachidonic acid from phospholipids , which is converted to PGE2 via cyclooxygenase 2 ( COX2 ) and membrane associated prostaglandin E synthase-1 ( mPGES-1 ) , the terminal enzyme that catalyzes the isomerization of prostaglandin H2 ( PGH2 ) to PGE2 [29 , 30] . PGE2-EP3 receptor signaling then leads to NLRC4 inflammasome assembly , which induces the release of IL-1β and IL-18 . Consistent with our previous reports where mice deficient in RIPK2 are susceptible to A . phagocytophilum infection [14] , we identified RIPK2 as a major regulator of the innate immune response against A . phagocytophilum . Ripk2-/- immune cells exhibited a defect in activation for the nuclear factor ( NF ) -κB and the NLRC4 inflammasome pathways . Altogether , we define the existence of a functionally distinct NLRC4 inflammasome upon microbial infection .
A . phagocytophilum transiently infects bone-marrow derived macrophages ( BMDMs ) [27 , 28] and clinical features in animal models and infected patients suggest classical macrophage activation [31–34] . To determine which genes are important for host immunity , we infected macrophages with A . phagocytophilum . Deep sequencing analysis [deposited at the Gene Expression Omnibus database ( GSE63647 ) ] indicated that the transcription of genes that encode for phospholipase A2 ( pla2g12a , pla2g5 and pla2g2e ) , COX2 ( ptgs2 ) and PGE synthase ( ptges ) was increased upon A . phagocytophilum infection ( Fig 1A ) . These genes are critical for prostanoid biosynthesis ( Fig 1B ) [35] and correlated with elevated enzymatic activities of cytosolic phospholipase A2 ( cPLA2 ) , COX1 and COX2 ( Fig 1C–1E ) , which led to increased levels of arachidonic acid ( AA ) , PGE2 , prostaglandin D2 ( PGD2 ) and thromboxane A2 ( TBXA2 ) ( Fig 1F–1I ) upon A . phagocytophilum infection . Eicosanoids have been associated with NLRC4 inflammasome activation [36] and phospholipase A2 releases arachidonic acid from phospholipids for eicosanoid biosynthesis ( Fig 1B ) [35] . Therefore , we examined whether cPLA2 was regulating the A . phagocytophilum-induced NLRC4 inflammasome . Pharmacological inhibition of cPLA2 , but not other phospholipases [e . g . , soluble phospholipase A2 ( sPLA2 ) , phospholipase C ( PLC ) and phospholipase D ( PLD ) ] reduced the levels of PGE2 , PGD2 and TBXA2 upon A . phagocytophilum infection of macrophages ( Fig 2A–2C ) . We also observed lower levels of IL-1β , IL-18 and caspase-1 activation upon bacterial stimulation of immune cells ( Fig 2D , 2E and 2G ) . Similar results were obtained with macrophages deficient in cPLA2 at low and high A . phagocytophilum multiplicity of infection ( MOI ) ( Fig 3A–3F and 3H ) , indicating that pharmacological inhibition of cPLA2 does not lead to off-target effects and the results obtained occurred independently of bacterial numbers . Importantly , secretion of IL-6 and translation of IL-1β and IL-18 by macrophages , which are not regulated by the inflammasome , remained unaffected during pre-treatment of macrophages with pharmacological inhibitors or in the absence of cPLA2 ( Fig 2F and 2G and Fig 3G and 3H ) . Surprisingly , chemical inhibition or genetic ablation of cPLA2 did not affect caspase-1 autoproteolysis and cytokine secretion when macrophages were infected with Salmonella ( S1 Fig ) , a pathogen that stimulates the NLRC4 inflammasome through the T3SS and flagellin [18–24] . Altogether , these results revealed that although both A . phagocytophilum and Salmonella trigger formation of the NLRC4 inflammasome , the signaling cascades that enable its activation appeared fundamentally different . To gain better insights into the A . phagocytophilum-induced NLRC4 inflammasome pathway , we pre-treated macrophages with the pan-COX inhibitor indomethacin [37] . Pre-treatment of cells with indomethacin followed by A . phagocytophilum infection decreased the release of PGE2 , PGD2 , TBXA2 , secretion of IL-1β and IL-18 , NLRC4 oligomerization and caspase-1 activation , but not IL-6 secretion by macrophages ( Fig 4 ) . To the contrary , pharmacological inhibition of lipoxygenase enzymes , 12/15-LOX ( PD146176 ) or 5-LOX ( AA861 ) , did not affect any of the parameters measured ( Fig 4 ) . Next , we pre-treated cells with celecoxib , a highly selective COX2 inhibitor [38] , followed by A . phagocytophilum infection . Pre-treatment of wildtype macrophages with celecoxib or , alternatively , A . phagocytophilum infection of COX2 ( Ptgs2 ) -deficient macrophages blunted the release of prostanoids , IL-1β and IL-18 , but not IL-6 secretion ( Fig 5A–5F and Fig 6A–6F ) . A . phagocytophilum infection of COX2 ( Ptgs2 ) -deficient macrophages and celecoxib inhibition of COX2 also decreased NLRC4 oligomerization and caspase-1 activation upon A . phagocytophilum infection ( Fig 5G and 5H and Fig 6G and 6H ) . As expected , no effect was observed for TLR4-deficient macrophages ( Fig 6 ) , as A . phagocytophilum does not carry genes for the biosynthesis of LPS in its genome [9] . Strikingly , Salmonella infection or nigericin stimulation of the NLRP3 inflammasome in COX2 ( Ptgs2 ) -deficient macrophages had no effect on the release of IL-1β , IL-18 , IL-6 , inflammasome oligomerization or caspase-1 activation ( S2B–S2F Fig ) . Secretion of PGE2 served as positive control for this experiment ( S2A Fig ) . The enzymatic activity of COX2 leads to the biosynthesis of prostanoids [38] . To determine which prostanoid affected the A . phagocytophilum-induced NLRC4 inflammasome , we performed a multi-pronged approach that included pharmacological inhibition , “add-back” assays and gene-targeted deletion of the membrane associated prostaglandin E synthase-1 ( mPGES-1 ) , the terminal enzyme that catalyzes the isomerization of PGH2 to PGE2 [29 , 30] . First , we observed that addition of PGE2 in macrophages deficient for ptgs2 ( COX2 ) restored caspase-1 function and IL-1β and IL-18 secretion upon A . phagocytophilum infection ( Fig 7A–7C ) . Conversely , the prostanoids PGD2 and TBXA2 did not elicit the activation of the NLRC4 inflammasome in the presence of A . phagocytophilum ( Fig 7A–7C ) . Second , specific pharmacological inhibition of the terminal PGE2 synthase enzyme , mPGES1 [29 , 30] , led to reduced caspase-1 activation and IL-1β and IL-18 secretion upon A . phagocytophilum infection in a dose-dependent manner ( Fig 7D–7H ) . Third , PGE2 “add-back” assays restored the phenotype in mPGES1-/- macrophages during A . phagocytophilum infection ( Fig 7I–7M ) . Importantly , secretion of IL-6 and translation of IL-1β and IL-18 by macrophages , which are not regulated by the inflammasome , remained unaffected during pharmacological inhibition , “add-back” and gene-targeted deletion assays ( Fig 7C , 7H and 7M and S3 Fig ) . Collectively , we provide convincing evidence that PGE2 is the sole eicosanoid that induces the activation of the NLRC4 inflammasome upon A . phagocytophilum infection . Next , we performed a kinetics experiment in macrophages to better characterize A . phagocytophilum infection in the context of NLRC4 inflammasome biology . As previously shown , A . phagocytophilum was undetectable inside macrophages at 2-hours post-infection [27] . A small number of bacteria was observed at 6 hours , followed by an increased load at 18 hours and reduction at 48 hours , which led to almost complete elimination after 72 hours of infection in macrophages ( Fig 8A ) [27] . Consistently , PGE2 secretion , caspase-1 activation and IL-1β and IL-18 secretion but not IL-6 , peaked at 18 hours , the same time point where the greatest number of A . phagocytophilum was detected inside macrophages ( Fig 8A–8D and S4A Fig ) . A . phagocytophilum does not synthesize LPS or peptidoglycans [8 , 9 , 11] . Therefore , one interesting immunological question pertains to the host molecule that induces NF-κB activation upon infection . We reasoned that RIPK2 could be this master regulator . This hypothesis rested on four findings . First , RIPK2 activates NF-κB signaling and mitogen activated protein ( MAP ) kinases upon infection [13] . Second , A . phagocytophilum interacts with the host endoplasmic reticulum ( ER ) [39] , which may exert RIPK2 activity in the absence of peptidoglycans due to cellular stress [40] . Third , COX2 expression is regulated through a signaling cascade that converges at the MAP kinase and the NF-κB pathways [41] . Fourth , mice deficient in RIPK2 are susceptible to A . phagocytophilum infection and secrete reduced levels of IL-18 in the peripheral blood [14] . Accordingly , ripk2-/- macrophages exhibited a defect in NF-κB and MAP kinase signaling , which led to decreased translation of COX2 , pro-IL-1β and IL-6 secretion ( Fig 8G and 8H ) . RIPK2 activity also affected PGE2 release and caspase-1 autoproteolysis upon A . phagocytophilum infection , as indicated by reduced levels of PGE2 , IL-1β , IL-18 and caspase-1 activation in cell culture supernatants of ripk2-/- macrophages ( Fig 8E , 8F and 8H and S4B Fig ) . Finally , A . phagocytophilum internalization was important for PGE2 release and NLRC4 inflammasome activation , as demonstrated in our experiments with cytochalasin D , a potent mycotoxin that inhibits actin polymerization ( Fig 8I–8K and S4C and S4D Fig ) . Collectively , we identified RIPK2 as a major regulator of the innate immune response against A . phagocytophilum . We then blunted the PGE2 signaling cascade with chemical antagonists that bind covalently to the four PGE2 receptor subtypes ( EP1-EP4 ) [30] and compared our findings with Salmonella . We observed that inhibition of the PGE2-EP3 receptor significantly decreased IL-1β and IL-18 release , and caspase-1 activation , but not IL-6 secretion upon A . phagocytophilum infection ( Fig 9C–9E and S5 Fig ) . The EP3 receptor for PGE2 is sensitive to pertussis toxin ( PT ) [42] . Macrophages pre-treated with PT and then stimulated with A . phagocytophilum also resulted in inhibition of the NLRC4 inflammasome ( S5A–S5D Fig ) . Importantly , the catalytically inactive pertussis toxin ( PT* ) , with a two amino acid substitution ( 9K129G ) [43] , did not block PGE2 signaling upon A . phagocytophilum colonization ( S5A–S5D Fig ) . Next , we took advantage of the ep3-/- mice and showed that in the absence of the EP3 receptor molecule , A . phagocytophilum did not induce caspase-1 activation and IL-1β and IL-18 secretion by macrophages ( Fig 9F–9I ) . Conversely , lack of the PGE2-EP3 receptor did not affect the NLRC4 inflammasome induced by Salmonella ( Fig 9J–9M and S6 Fig ) . PGE2 exerts its actions by acting on G-protein-coupled receptors ( GPCRs ) . PGE2 binds to the EP3 receptor , which inhibits the membrane associated adenylyl cyclase via Gαi [44] . This signaling relay decreases cytosolic cyclic AMP ( cAMP ) production , as adenylyl cyclase catalyzes the conversion of adenosine triphosphate ( ATP ) to cAMP [44] ( S8 Fig ) . We validated these observations with sulprostone , an EP3 agonist and positive control in our assays ( S7A Fig ) . Consistently , A . phagocytophilum colonization of macrophages led to reduced production of cAMP ( S7A Fig ) . Moreover , pharmacological blockade of the PGE2-EP3 receptor via the EP3 antagonist or PT hindered the inhibition of cAMP by A . phagocytophilum ( S7A Fig ) . Next , we showed that membrane , but not soluble , adenylyl cyclase modulated the A . phagocytophilum-induced NLRC4 inflammasome . Forskolin , a selective inhibitor of the membrane-associated adenylyl cyclase [45] , inhibited IL-1β , IL-18 and caspase-1 autoproteolysis during A . phagocytophilum infection of macrophages ( S7B–S7E Fig ) . On the other hand , pre-treatment of macrophages with KH7 , a specific pharmacological inhibitor of soluble adenylyl cyclase [45] , did not affect NLRC4 inflammasome function during A . phagocytophilum infection ( S7F–S7I Fig ) . Altogether , these findings: ( i ) indicated that the PGE2-EP3 axis is critical for the NLRC4 inflammasome elicited by A . phagocytophilum; and ( ii ) explained why Salmonella is unable to trigger a similar pathway when compared to A . phagocytophilum . This was likely due to reduced expression of the EP3 receptor during Salmonella infection of macrophages ( Fig 9A and 9B ) . To prove that the results obtained in vitro could also be observed in vivo , we then infected mice deficient in COX2 ( Ptgs2 ) with A . phagocytophilum . Ptgs2-deficient animals were more susceptible to A . phagocytophilum infection ( Fig 10A ) and exhibited reduced levels of IL-18 in the peripheral blood when compared to the wildtype mice ( Fig 10B ) . As previously seen , no detectable levels of IL-1β were observed in the blood of A . phagocytophilum-infected mice [25] . These findings agreed with our prior publications , showing that IL-18 release mediated by RIPK2 and the NLRC4 inflammasome regulates interferon ( IFN ) -γ production by CD4+ T cells upon A . phagocytophilum infection [14 , 25] . COX2 ( Ptgs2 ) -deficient mice infected with A . phagocytophilum also revealed lower levels of PGE2 , PGD2 , TBXA2 and splenomegaly ( Fig 10C–10G ) . COX2 ( Ptgs2 ) -deficient animals had increased cellular infiltration in the red pulp and damage to the splenic architecture upon A . phagocytophilum infection ( Fig 10H ) . In sum , these results showed that COX2 is critically important for A . phagocytophilum infection in vivo .
The NLRC4 inflammasome is currently thought to only recognize components of the bacterial T3SS and flagellin [18–24] . Other inflammasomes , however , such as the NLRP3 scaffold , sense a wide-range of molecular structures leading to caspase-1 activation and cytokine secretion [16] . We hypothesized that an alternative signaling cascade for the NLRC4 inflammasome must exist because mice deficient in NLRC4 are susceptible to A . phagocytophilum infection [25] , an obligate intracellular rickettsial bacterium that does not have a T3SS and flagellin [11] . Furthermore , Annexin a2-deficient mice were more susceptible to A . phagocytophilum infection and showed splenomegaly , thrombocytopenia and monopenia [28] . Macrophages deficient in Annexin A2 , a phospholipid-binding protein , secreted significantly smaller amounts of IL-1β and IL-18 and had a defect in NLRC4 inflammasome oligomerization and caspase-1 activation [28] . In contrast , Annexin a2-/- macrophages released IL-1β , IL-18 , and IL-6 at wild-type levels when infected with Salmonella , a canonical NLRC4 agonist [28] . We provide unequivocal evidence that two distinct signaling pathways occur for NLRC4 inflammasome activation within the cell: one termed classical ( i . e . , stimulated by Salmonella ) and another referred to as alternative ( i . e . described here , responding to A . phagocytophilum ) . Given how inflammasome biology intersects with a growing number of disciplines , we reason that these findings are conceptually valuable because we reveal that eicosanoid receptors in immune cells activate diverging signaling cascades . For instance , both A . phagocytophilum and Salmonella lead to PGE2 production by macrophages . However , Salmonella is unable to activate the eicosanoid-dependent NLRC4 inflammasome pathway because it does not induce PGE2-EP3 receptor expression . PGE2 is likely acting in an autocrine/paracrine manner to drive NLRC4 inflammasome activation upon A . phagocytophilum infection . This is based on the evidence that A . phagocytophilum infection upregulates the EP3 receptor , which is known to elicit PGE2 signaling in a cell-intrinsic manner [30 , 35] . Alternatively , PGE2 may also affect the function of “bystander” cells in a paracrine manner given that our exogenous PGE2 “add-back” assays restored NLRC4 inflammasome activity in A . phagocytophilum-infected cells . Can rickettsial agents be used to uncover broader principles of immune sensing ? The answer to this question may have to deal with the biology of these organisms . Rickettsial agents differ greatly in terms of how they invade and replicate within the mammalian host when compared to other bacteria commonly used to study microbial immunity . Their obligate intracellular life style , coupled to the intense selective pressure to survive both in the arthropod vector and the mammalian host [1 , 2] suggests that these microbes have to employ extreme measures to conceal themselves from the immune system . This reasoning may explain why A . phagocytophilum triggers such a distinct pathogen-recognition mechanism when compared to other bacteria . In summary , we discovered a novel mode of NLRC4 inflammasome activation triggered by the rickettsial bacterium A . phagocytophilum . We revealed that some microbial pathogens lacking the T3SS and flagellin activate the NLRC4 inflammasome . We also illustrated how this protein scaffold distinguishes bacterial infection within the cell . Altogether , our findings suggest that there are broader yet-to-be discovered principles of microbial sensing in the context of NLRC4 inflammasome biology .
Breeding and experiments were performed in strict compliance with guidelines set forth by the National Institutes of Health ( Office of Laboratory Animal Welfare [OLAW] assurance number A3200-01 ) . Procedures were approved by the Institutional Biosafety ( IBC:00002247 ) and Animal Care and Use ( IACUC:0413017 and 0216015 ) committees at the University of Maryland , Baltimore . Ripk2-/- ( 007017 ) , C57BL/6 ( 000664 ) and Ptgs2-/- ( COX2 ) mice ( 008101 ) were purchased from Jackson Laboratories . Femurs from mPGES1-/- [29] and Ep3-/- [46] mice were a gift from Leslie Crofford and Richard Breyer at Vanderbilt University School of Medicine . Tlr4-/- and cPla2-/- mice were previously described [47 , 48] . Mice were gender matched and at least 6–10 weeks of age . BMDMs were generated , as previously described [27] . Culturing for the A . phagocytophilum strain HZ and calculations were described elsewhere [27] . Salmonella strain SL1344 was a gift from Dr . Stefanie Vogel at the University of Maryland , Baltimore School of Medicine . Salmonella was grown in HS media at 37°C and enumerated , as previously described [49] . Cell cultures were tested and determined to be Mycoplasma-negative through a commercially available PCR kit ( Southern Biotech -13100-01 ) . LPS ( 50ng/ml ) was purchased from InvivoGen . Nigericin ( 10μM ) , indomethacin ( 100 nM ) and celecoxib ( 0 . 1μM to 10 μM ) were purchased from Sigma-Aldrich . AA861 ( 1μg/ml ) and PD146176 ( 1μg/ml ) were purchased from BioMol International . CAY10526 ( 10010088 ) , KH7 ( 13243 ) , Forskolin ( 11018 ) , Cytochalasin D ( 11330 ) , PGE2 ( 14010 ) , PGD2 ( 12010 ) and U46619 ( thromboxane A2 analogue , 16450 ) were purchased from Cayman Chemicals . The PGE2 receptor antagonists EP1–1μM ( SC51089 ) , EP2–5μM ( AH6809 ) and EP4–5μM ( ONO-AE3-208 ) were purchased from Cayman Chemical , whereas the antagonist for the PGE2 EP3–10μM ( L-798106 ) and the PGE2 EP3 receptor agonist ( sulprostone—3μM ) was purchased from Sigma . The inhibitors for the phospholipases cPLA2 ( AACOCF3 ) , sPLA2 ( LY315920 ) , PLC ( U73122 ) and PLD ( FIPI ) were purchased from Tocris Bioscience . Pertussis toxin ( PT ) and the catalytically inactive pertussis toxin ( PT* ) with a two amino acid substitution ( 9K129G ) were described previously [43 , 50] . 1×106 BMDMs were seeded into 24-well plate in 300 μl of media containing 5% fetal bovine serum ( FBS ) overnight prior to the challenge by either A . phagocytophilum ( MOI 10 and 50 ) or Salmonella ( MOI 25 ) for 1 hour . 50ng/ml of LPS was used for cell priming at 37°C and 5% CO2 for 30 minutes during Salmonella infection . LPS-primed cells were washed twice extensively followed by the addition of bacteria . In inhibition assays , 1×106 WT and genotype-deficient BMDMs were pre-treated with pharmacological inhibitors at indicated time and concentrations followed by the stimulation with A . phagocytophilum ( MOI 10 and 50 ) overnight or Salmonella ( MOI 25 ) for 1 hour . For the Ptgs2-/- and mPGES1-/- “add-back” experiments , 1×106 WT and deficient cells were infected with A . phagocytophilum ( MOI 50 ) for 4 hours followed by the addition of the respective eicosanoid at indicated concentrations for 18 hours . After infection , cultured supernatants and cell lysates collected from each well were used for ELISA and immunoblot assays . Equal amounts of supernatants were mixed with the native sample buffer ( 62 . 5 mM Tris-HCl , 40% glycerol , 0 . 01% bromophenol blue , pH 6 . 8 ) , loaded into 4–15% Mini-PROTEAN TGX Precast Gels and run at 200 volts for 2 hours in 1×Tris/Glycine native running buffer ( 25 mM Tris , 192 mM glycine , pH8 . 3 ) . NativeMark Unstained Protein Standard ( Invitrogen ) was visualized with Gel Code Blue Safe Protein Stain solution ( Thermo Scientific ) . Cell lysates were prepared in radioimmunoprecipitation ( RIPA ) lysis buffer ( Boston Bioproducts ) with Halt Protease Inhibitor Cocktail ( Thermo Scientific ) and PhosSTOP ( Roche Applied Science ) . 4–15% Mini-PROTEAN TGX precast gels were run at 200 volts for 30 minutes in the 1×Tris-Glycine-SDS running buffer ( Boston Bioproducts ) . Transfer was performed using the Bio-Rad Trans-Blot Turbo with either polyvinylidene fluoride ( PVDF ) or nitrocellulose membranes ( Bio-Rad ) . Membranes were blocked in 5% skim milk or BSA ( Bio-Rad ) . Western blot antibodies for caspase-1 ( 1:1000 , Millipore 06–503 or 06-503-I , 1:1 , 000 , Proteintech 22915-1-AP; 1:2000 Genentech 4175 , cell line 4B4 . 2 . 1 , or 1:1000 , AdiPoGen International AG-20B-0042 ) , NLRC4 ( 1:1000 , Millipore , 06–1125 ) , IL-1β ( 1:1000 R&D Systems and Cell Signaling , AF401-NA and 12426S ) , IL-18 ( 1:1000 , MBL JM-5180-100 ) , β-actin ( 1:1000 , Sigma A2103 ) , COX2 ( 1:1000 , Cell Signaling 12282 ) , phospho-IκB-α ( 1:1000 , Cell Signaling 9246s ) , p-ERK ( 1:400 , Cell Signaling 4370 ) , ERK ( 1:1000 , Cell Signaling 9102 ) , IκB-α ( 1:1000 , Cell Signaling 4812 ) , PTGER3 ( 1:1 , 000 , Abcam ab117998 ) , anti-mouse horseradish peroxidase ( HRP ) , anti-goat HRP , anti-rabbit HRP ( 1:5000 , Abcam ab97046 , ab97110 and ab97051 , respectively ) , anti-rat HRP ( 1:5000 Abcam and Santa Cruz Biotechnology , ab97057 and sc-2006 ) were used . A rabbit polyclonal antibody raised against A . phagocytophilum [51] was kindly provided by Erol Fikrig at Yale University School of Medicine ( 1:2 , 000 ) . Enhanced chemiluminescence ( ECL ) western blotting substrate and Super Signal West Pico Chemiluminescent substrate were used ( Thermo Scientific ) . Restore Western Blot Stripping Buffer was used for the stripping of antibodies on the blots ( Thermo Scientific ) . IL-1β and IL-6 were measured with the BD OptEIA Set ( BD Biosciences ) . IL-18 capture ( 1:1 , 000 , D047-3 ) and detection antibodies ( 1:2 , 000 , D048-6 ) were purchased from MBL . PGE2 was measured with the ELISA kit ( Enzo Life Sciences ) . PGD2 was measured with the ELISA kit ( Cayman Chemicals ) . Thromboxane A2 was measured with the Mouse Thromboxane A2 ELISA Kit ( Abbexa ) . Quantitate RT-PCR was performed using the Power SYBR Green PCR Master Mix ( Invitrogen ) in an ABI 7500 real-time PCR instrument . Primer sequences for A . phagocytophilum were as follows: 16S-F ( 5’-CAGCCACACTGGAACTGAGA-3’ ) and 16S-R ( 5’-CCCTAAGGCCTTCCTCACTC-3’ ) . Gene expression was normalized by using the primers β-actin-F ( 5’-ACGCAGAGGGAAATCGTGCGTGAC-3’ ) and β-actin-R ( 5’-ACGCGGGAGGAAGAGGATGCGGCAGTG-3’ ) . The absolute quantification method was used . For the PGE2-EP3 receptor quantification , PureLink RNA Mini Kit ( Invitrogen ) and the Verso cDNA synthesis Kit ( Thermo Scientific ) were used . Gene expression was normalized by using the primers GAPDH-F ( 5’-TGATGACATCAAGAAGGTGGTGAAG-3’ ) and GAPDH-R ( 5’-TCCTTGGAGGCCATGTGGGCCAT-3’ ) . Primer sequences for the EP3 receptor were as follows: EP3-F ( 5’-GGTTCCTGTGAAGGACTGAAGAC-3’ ) and EP3-R ( 5’-AAGGTTCTGAGGCTGGAGATA-3’ ) . The relative quantification method ( fold changes ) was used . 15×106 wildtype cells were stimulated with A . phagocytophilum ( MOI 25 ) overnight . Cells were scraped followed by sonication . COX1/2 enzymatic assays were performed with COX activity assay kit ( Cayman Chemicals ) , whereas cPLA2 activity was measured following instructions by the manufacturer ( Abnova ) . Arachidonic acid levels were measured according to the instructions of the ELISA kit ( MyBiosource ) . cAMP was measured by using the cyclic AMP XP Assay Kit ( Cell Signaling Technology ) . BMDMs were grown into 6-well culture plates at 7×106 per well . Cells were stimulated with A . phagocytophilum . Uninfected BMDMs were used as controls and the experiment was performed in triplicate . Total RNA was isolated with the PureLink RNA Mini Kit ( Invitrogen ) . Illumina Sequencing was performed at the University of Maryland , Baltimore . Briefly , Illumina RNAseq libraries were prepared with the TruSeq RNA Sample Prep kit ( Illumina , San Diego , CA ) . The indexed libraries were pooled and sequenced using the HiSeq platform ( Illumina ) for the mouse samples in order to generate 101 base pair reads . The reads were further trimmed due to low quality at the trailing 3' end . These trimmed paired end reads were populated into 2 separate FASTQ format files and the quality of the reads was tested using the FastQC toolkit to ensure quality of the sequencing reads . The RNA sequencing reads were used as input for the TopHat read alignment tool to be aligned to the mouse genomic reference sequence ( Ensembl GRCm38 version ) for each of the samples . The reference genomic sequences for the GRCm38 genome build were downloaded from the Ensembl resources . The output from TopHat was obtained as BAM format files . In the alignment phase , we allowed up to two mismatches per 30 base pair segment and removed reads that aligned to more than 20 genomic locations . The BAM alignment files obtained from the TopHat alignment tool was analyzed to generate the alignment statistics for each sample , namely , the total number of reads , the number of mapped reads and the percent of mapped reads . For the differential gene expression analysis , the alignment BAM files from TopHat were further utilized to compute gene expression levels and test each gene for differential expression . The mouse gene set reference annotation ( version GRCm38 ) in GTF format was downloaded from the Ensembl resources . The number of reads that mapped to each gene described in the Ensembl annotation was calculated using the python package HTSeq-an alignment read count tool . The read count represented the expression of the gene . Differential gene expression analysis was conducted using the DESeq R package ( available from Bioconductor ) . The DESeq analysis resulted in the determination of differentially expressed genes . DESeq utilized the read counts provided by the HTSeq read count tool . The read counts for each sample were normalized for sequencing depth and distortion caused by highly differentially expressed genes . The negative binomial model was used to test the significance of differential expression between two genotypes . The differentially expressed genes were deemed significant if the FDR ( False Discovery Rate ) was less than 0 . 01 , the gene expression was above the 45th percentile and gene showed greater than 2-fold change difference ( over expressed or under expressed ) between conditions . Principal component analysis and other clustering methods were used to visualize the clustering of the replicates across samples . Heat maps were generated to illustrate the genes showing significant differences between multiple comparisons of the control and other infection and/or treatment conditions . C57BL/6 ( n = 20 ) and COX2 ( Ptgs2 ) -/- ( n = 10 ) mice were infected by intraperitoneal injection with A . phagocytophilum strain HZ ( 1×107 cells ) . Blood samples were collected at days 0 , 5 and 10 for the IL-18 ELISA . Spleens were removed , normalized to the body weight , and compared to those of non-infected mice . Spleens were fixed at day 15 post-infection with 10% neutral buffered formalin and embedded in paraffin wax . Sections ( 5 μm ) were obtained and stained with hematoxylin and eosin . Measurement of A . phagocytophilum load was done at day 15 post-infection in the peripheral blood of infected animals using quantitative RT-PCR , as described above . All experiments in this study were performed with at least 2–5 replicates . All data were expressed as means ± standard errors of the means ( SEM ) . The differences between groups were examined by either unpaired Student's t test or one-way analysis of variance ( ANOVA ) . All statistical calculations and graphs were made by using GraphPad Prism version 6 . 0 . P < 0 . 05 was considered statistically significant . | Elimination of bacteria is orchestrated by the immune system . Intracellular bacteria are generally recognized by cytosolic molecules named Nod-like receptors ( NLRs ) . One such protein scaffold that senses needle-like structures and globular proteins , namely , the bacterial type III secretion ( T3SS ) and flagellin , is the NLRC4 inflammasome . The NLRC4 inflammasome induces caspase-1 autoproteolysis and secretion of the pro-inflammatory cytokines interleukin ( IL ) -1β and IL-18 . Here , we show that the obligate intracellular rickettsial pathogen Anaplasma phagocytophilum , which does not have a T3SS or flagellin-coding genes , induces a distinct NLRC4 inflammasome circuitry through the eicosanoid prostaglandin E2 and the EP3 receptor . Conceptually , these findings establish the existence of a distinct microbial surveillance system where the NLRC4 inflammasome senses an obligate intracellular pathogen of public health relevance . Therefore , we propose that rickettsial agents can be used to uncover broader principles of immune surveillance given their unique life style to survive inside the mammalian host and lack of pathogen associated molecular patterns commonly present in most facultative intracellular bacteria . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"pathogens",
"immunology",
"biological",
"cultures",
"microbiology",
"immunoblotting",
"salmonellosis",
"bacterial",
"diseases",
"physiological",
"processes",
"enterobacteriaceae",
"inflammasomes",
"cell",
"cultures",
"molecular",
"biology",
"techniques",
"immunologic",
"techniques",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"immunoassays",
"microbial",
"pathogens",
"molecular",
"biology",
"salmonella",
"biochemistry",
"cell",
"biology",
"physiology",
"secretion",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2016 | The Prostaglandin E2-EP3 Receptor Axis Regulates Anaplasma phagocytophilum-Mediated NLRC4 Inflammasome Activation |
Vaccinia virus ( VACV ) is the prototypic orthopoxvirus and the vaccine used to eradicate smallpox . Here we show that VACV strain Western Reserve protein 169 is a cytoplasmic polypeptide expressed early during infection that is excluded from virus factories and inhibits the initiation of cap-dependent and cap-independent translation . Ectopic expression of protein 169 causes the accumulation of 80S ribosomes , a reduction of polysomes , and inhibition of protein expression deriving from activation of multiple innate immune signaling pathways . A virus lacking 169 ( vΔ169 ) replicates and spreads normally in cell culture but is more virulent than parental and revertant control viruses in intranasal and intradermal murine models of infection . Intranasal infection by vΔ169 caused increased pro-inflammatory cytokines and chemokines , infiltration of pulmonary leukocytes , and lung weight . These alterations in innate immunity resulted in a stronger CD8+ T-cell memory response and better protection against virus challenge . This work illustrates how inhibition of host protein synthesis can be a strategy for virus suppression of innate and adaptive immunity .
The study of virus-host interactions continues to provide valuable information about the complex relationships between cells and pathogens . Large DNA viruses , in particular , encode many proteins that modify the intracellular environment to promote viral survival , replication and spread . Vaccinia virus ( VACV ) is the prototypic Orthopoxvirus of the Poxviridae and is the vaccine used to eradicate smallpox [1] . VACV replicates in the cytoplasm and encodes about 200 proteins that are required for viral transcription and replication [2 , 3] , alteration of cell metabolism [4–7] , and immune evasion [8] . Between one-third and one-half of VACV proteins are devoted to evasion of innate immunity and these immunevasins may function inside or outside the infected cell . Intracellular immunevasins include those that inhibit innate immune signaling pathways leading to activation of nuclear factor kappa-light-chain-enhancer of activated B cells ( NF-κB ) , interferon ( IFN ) regulatory factor ( IRF ) -3 and Janus kinase ( JAK ) / signal transducer and activation of transcription ( STAT ) signaling . Other intracellular immunevasins suppress apoptosis or the antiviral activity of IFN-stimulated gene products . Additional immunevasins are secreted from infected cells to bind complement factors , IFNs , cytokines or chemokines extracellularly and inhibit their activity . An interesting aspect of these immune evasion strategies is the apparent redundancy , with several proteins targeting the same activation pathway . For instance , there are at least 10 intracellular inhibitors of NF-κB encoded by VACV [9–18] and a VACV strain lacking all these factors still inhibits NF-κB [19] . VACV , like all viruses , relies on host ribosomes for virus protein synthesis . To ensure efficient translation of virus proteins , VACV shuts off host protein synthesis and re-directs the cellular translational machinery to the synthesis of viral proteins [20–27] . VACV mRNAs are translated by a cap-dependent mechanism facilitated by the eukaryotic initiation factor ( eIF ) 4F complex that recognizes the 5’-methylated cap , and translation is initiated by interaction of the cap with eIF4E , a cap-binding protein [28] . VACV encodes capping [29] and methylating enzymes [30] that produce viral mRNAs that mimic cellular mRNAs and so evade detection by host pattern recognition receptors . VACV protein synthesis occurs in virus factories [21 , 27 , 31] , and to ensure preferential translation of virus mRNAs , VACV expresses de-capping enzymes D9 and D10 that remove the cap from both cellular and viral mRNAs [25 , 32 , 33] . The abundance of viral transcripts ensures translation of viral mRNA continues despite this de-capping activity , which also promotes turnover of viral mRNAs and thereby aids the transition between the early , intermediate and late stages of viral gene expression . The importance of protein D10 for the virus replication cycle is illustrated by a D10 deletion mutant that has a smaller plaque phenotype and produces reduced yields of virus in cell culture [26] . Moreover , mutant viruses with a stop codon introduced into the D10 open reading frame ( ORF ) or with amino acid alterations in the D10 catalytic site have an attenuated phenotype in vivo [34] . D9 and D10 also reduce dsRNA accumulation and the consequential activation of host responses [35] . A similar outcome was observed after VACV infection of cells lacking the host exonuclease Xrn1 [36] . This report presents a functional characterization of VACV strain Western Reserve ( WR ) protein 169 , a previously uncharacterized protein that is expressed by some , but not all VACV strains and orthopoxviruses . Protein 169 is an inhibitor of cap-dependent and cap-independent translational initiation . Protein 169 localizes in cytoplasmic puncta and is largely excluded from virus factories , enabling preferential inhibition of host mRNA translation . Consistent with this , protein 169 does not affect virus replication or spread in cell culture , but is a potent inhibitor of translation in cells in which it is expressed ectopically . Consequently , protein 169 blocks expression of host proteins that are induced following activation of diverse innate immune signaling pathways , and , in two in vivo models of VACV infection , a virus lacking 169 ( vΔ169 ) induces a more severe primary infection than control viruses . The altered disease severity is not due to changes in viral replication , but instead is associated with increased production of pro-inflammatory cytokines and chemokines , and increased recruitment of immune cells at the site of infection . This altered response also affects the adaptive memory response and causes increased CD8+ T-cell memory and better protection against virus challenge . Collectively , these results indicate that virus inhibition of host protein synthesis can be a strategy to suppress innate and adaptive immunity , rather than primarily a means to aid virus replication as considered hitherto .
VACV strain WR gene 169R encodes a small , charged protein of 78 amino acid residues . The protein lacks a nuclear localisation signal and a hydrophobic transmembrane sequence suggesting that protein 169 is likely to be cytosolic . The ORF is conserved in VACV strains modified vaccinia virus Ankara ( MVA ) , Lister , Duke , Acambis 3000 and rabbitpox virus , and other orthopoxviruses such as camelpox virus , taterapox virus , cowpox virus and monkeypox virus ( S1 Fig ) . However , the ORF is truncated in multiple variola virus strains ( the cause of smallpox ) after codon 38 and in ectromelia virus ( ECTV ) after codon 41 . In cowpox virus and monkeypox virus there are minor changes in amino acid length and composition , but the protein is identical in the VACV strains shown and in taterapox virus ( S1 Fig ) . The truncation of this ORF in VACV strain Copenhagen and in other orthopoxviruses indicates that the 78 amino acid protein is non-essential for orthopoxvirus replication . The expression of protein 169 by several VACV strains was investigated by immunoblotting using a rabbit polyclonal antibody raised against VACV WR protein 169 that had been expressed in and purified from E . coli ( Methods ) . This detected a 13-kDa polypeptide in cells infected with VACV strains WR , MVA , Lister , rabbitpox , International Health Department ( IHD ) -J and Tian Tan , and cowpox virus strain Brighton Red , but not VACV strain Copenhagen , or in mock-infected cells ( Fig 1A ) . VACV infection was confirmed by immunoblotting with a mAb that recognizes the VACV structural protein D8 [37] , although this mAb did not detect the D8 protein made by MVA ( Fig 1A ) . Immunoblotting for α-tubulin demonstrated equal loading of samples . The time of expression and localization of protein 169 during infection were investigated by immunoblotting ( Fig 1B and 1C ) and immunofluorescence microscopy ( Fig 1D ) . HeLa cells were infected with v169 ( a plaque purified , wild-type virus that expresses protein 169 ) in the presence or absence of cytosine arabinoside ( AraC ) , a DNA replication inhibitor that blocks intermediate and late VACV gene expression . The anti-169 antiserum detected a 13-kDa protein from 2 h p . i . that was also present following addition of AraC , showing expression prior to DNA replication ( Fig 1B ) . Similar expression kinetics were observed for early VACV protein C16 [38] . In contrast , the VACV late protein D8 [39] was expressed only in the absence of AraC . The localization of protein 169 was investigated by biochemical fractionation of infected cells . Immunoblotting of lysates from cells infected with v169 , vΔ169 ( a deletion mutant lacking the 169R gene ) and v169-rev ( a revertant virus in which the 169R gene was reinserted at its natural locus into vΔ169 ) showed that protein 169 is expressed from v169 and v169-rev , but not vΔ169 , and that it localizes predominantly in the cytoplasm . Satisfactory separation of cytoplasmic and nuclear fractions was confirmed by blotting for α-tubulin and lamin ( Fig 1C ) . Analysis by immunofluorescence using purified anti-169 antibody ( Methods ) detected protein 169 from 4 h p . i . in cytoplasmic puncta ( Fig 1D ) . VACV factories were also detected from 4 h p . i . by DAPI staining , but protein 169 was excluded from these structures . To determine if protein 169 co-localized with specific cytoplasmic organelles , infected cells were stained with antibodies that detected the endoplasmic reticulum , mitochondria , Golgi apparatus , clathrin-containing vesicles and endosomes but no clear co-localization was observed ( Fig 2A ) . Partial co-localization with 40S ribosomes was noted , although the abundance of 40S ribosomes makes a clear correlation uncertain . Staining with DAPI confirmed that protein 169 was excluded from virus factories ( Fig 2B ) . The contribution of protein 169 to virus replication and spread was investigated using recombinant VACVs v169 , vΔ169 , and v169-rev that were constructed by transient dominant selection [40] ( Methods ) . These three viruses formed plaques of indistinguishable size in African monkey fibroblasts ( BSC-1 ) and also in rabbit kidney ( RK ) -13 cells and human TK-143 cells ( Fig 3A–3C ) . Similarly , the yields of intracellular and extracellular vΔ169 were unaltered compared to control viruses after high ( 10 PFU/cell ) or low ( 0 . 05 PFU/cell ) multiplicity of infection in BSC-1 cells ( Fig 3D–3G ) . Therefore , the 169 protein is non-essential for virus replication and spread in cell culture . The 169R gene is located in a terminal variable region of the VACV genome , is expressed early during infection and is non-essential for virus replication in cell culture . These properties are characteristic of VACV genes encoding immunevasins , such as the type I IFN binding protein [41 , 42] , the 3-β-hydroxysteroid dehydrogenase [43 , 44] and the intracellular inhibitors of NF-κB activation [9 , 10 , 12–18] . Therefore , we hypothesized that protein 169 might be an immunevasin and this was tested by reporter gene assays . A plasmid in which firefly luciferase expression is driven by either an NF-κB , IRF-3 ( ISG56 . 1 ) , or interferon-stimulated response element ( ISRE ) responsive promoter was transfected separately into HEK 293T cells together with TK renilla luciferase ( internal control ) , and plasmids expressing 169 , FLAG-tagged 169 ( FLAG-169 ) or other control proteins . The controls chosen were ( i ) VACV strain WR protein B14 that inhibits the NF-κB signaling by binding to IKKβ [15] , ( ii ) VACV protein C6 that inhibits IRF-3 signaling by binding to TBK-1 adaptors [45] , and ( iii ) paramyxovirus protein PiV5-V that inhibits type I IFN-induced signaling by degrading STAT1 [46] . Luciferase activity was measured by luminescence after stimulation with TNF-α ( NF-κB Luc ) , IFN-α ( ISRE Luc ) or after transfection with poly ( I:C ) ( IRF-3 Luc ) . Protein 169 and FLAG-169 inhibited NF-κB , IRF-3 and ISRE pathways as well as , or better than , known inhibitors of these pathways ( Fig 4A–4C ) . The inhibition of all these pathways was surprising , and contrasted with the controls that generally inhibit specific pathways only . Interestingly , protein 169 also caused reduced expression of TK renilla , suggesting a general reduction in protein expression . To investigate this further , the levels of chemokine CXCL10 were measured by ELISA . HEK 293T cells were transfected with plasmids expressing GFP , VACV B14 , C6 , 169 or Δ12A49 and then infected with Sendai virus ( SeV ) . VACV protein A49 inhibits NF-κB signaling by binding to the E3 ubiquitin ligase β-TrCP but deletion of the first 12 amino acids abolishes this function [17] and so Δ12A49 served as a negative control . After 24 h , CXCL10 in the supernatant was measured by ELISA ( Fig 4D ) . CXCL10 expression is induced by both NF-κB and IRF-3 , and so levels of CXCL10 were lower in cells expressing either B14 or C6 , but not in cells expressing Δ12A49 , as expected . However , protein 169 also reduced CXCL10 levels , consistent with results of the reporter gene assays . To test whether 169 mediates its inhibitory activity by blocking transcription , the levels of specific mRNAs were measured . A549 cells were transfected with plasmids expressing GFP , VACV B14 , C6 , or 169 and were stimulated 24 h later with TNF-α . mRNA levels of NF-κB-inducible genes such as intercellular adhesion molecule 1 ( ICAM-1 ) , IL-6 , and NFκBia were measured by reverse transcription quantitative-PCR ( RT-q-PCR ) and normalized to the housekeeping gene hypoxanthine-guanine phosphoribosyltransferase ( HPRT ) ( Fig 4E–4G ) . Levels of all three mRNAs were similar in cells expressing 169 , GFP , or C6 following stimulation with TNF-α . Conversely , as expected , lower levels of these NF-κB-inducible mRNAs were detected in cells expressing the NF-κB inhibitor B14 . No difference was detected in HPRT mRNA levels , confirming that the 169-mediated inhibition of multiple immune signaling pathways was not due to a general inhibition of transcription . Therefore , it was likely that protein 169 inhibited gene expression either by blocking mRNA transport to the cytoplasm , or by blocking protein synthesis . The former possibility was unlikely given that protein 169 is cytoplasmic , but was addressed by measuring the levels of cytoplasmic and nuclear mRNAs . HEK 293T cells were co-transfected with plasmids expressing NEMO fused with renilla luciferase ( NEMO-Luc ) and protein 169 . A plasmid expressing protein A49 and an empty vector were included as negative controls and cycloheximide was added as an inhibitor of translation . The levels of luciferase-tagged proteins were determined by luminescence ( S2A Fig ) and mRNA levels of NEMO-Luc were determined by RT-q-PCR ( S2B Fig ) . In parallel , cytoplasmic and nuclear mRNAs were extracted and mRNAs levels of NEMO-Luc , HPRT and TATA box-binding protein were compared in these fractions ( S2D–S2F Fig ) . As before , only low levels of NEMO-Luc was detected in cells expressing 169 or treated with cycloheximide . Slightly lower cytoplasmic mRNA levels of NEMO-Luc were found in cells expressing 169 , but this slight decrease could not explain the profound ( ~10-fold ) reduction of NEMO-Luc . There was also a slight reduction in NEMO-Luc mRNA in the cytoplasm in cycloheximide-treated cells , suggesting such reduction might derive from a general inhibition in protein synthesis . Lastly , no decrease in endogenous mRNAs was observed in the presence of protein 169 . Collectively these data indicate that mRNA transcription and export are not inhibited by protein 169 and therefore its inhibitory effect is downstream . To investigate if protein 169 inhibits protein synthesis , HeLa cells were co-transfected with plasmids expressing GFP together with VACV N1 , 169 , FLAG-169 or empty vector . VACV N1 is another inhibitor of NF-κB signaling [47 , 48] and served as a negative control . GFP levels were determined by immunoblotting and GFP mRNAs were measured by RT-q-PCR ( Fig 5A and 5B ) . Cycloheximide , 169 and FLAG-169 reduced GFP levels greatly compared with N1 or empty vector . In contrast , GFP mRNA levels were similar in all cells and were higher in cells treated with cycloheximide . These experiments reveal that protein 169 inhibited protein synthesis and that this is generic rather than being specific to proteins functioning in innate immunity . VACV inhibits cap-dependent translation of host mRNAs by the de-capping enzymes D9 and D10 , but these do not affect cap-independent translation [25] . To determine if protein 169 has similar or different specificity , its ability to inhibit cap-dependent and internal ribosome entry site ( IRES ) -dependent translation was evaluated . A plasmid encoding a bicistronic gene in which firefly luciferase is translated in a cap-dependent manner and renilla luciferase is translated in a foot and mouth disease virus ( FMDV ) IRES-dependent manner was transfected into HEK 293T cells together with 169 , FLAG-169 or 169-AAG . The latter plasmid has the 169 initiation codon and the fourth codon mutated from AUG to AAG to prevent translation and distinguish between inhibition mediated by 169 mRNA or 169 protein . Luciferase levels were determined by luminescence ( Fig 5C and 5D ) , mRNA levels were determined by RT-q-PCR ( Fig 5E and 5F ) and protein expression was also measured by immunoblotting ( Fig 5G ) . Low levels of both firefly and renilla luciferase were found in the presence of cycloheximide , 169 and FLAG-169 , but not 169 AAG , confirming that the inhibitory effect of 169 on translation requires protein 169 . In contrast , luciferase levels were unaffected by proteins N1 or A49 . Similar mRNA levels of renilla luciferase were found in all samples . These data show that protein 169 inhibits both cap-dependent and FMDV IRES-dependent translation . To evaluate the influence of protein 169 on protein synthesis in uninfected cells and during VACV infection , nascent proteins were analysed using surface sensing of translation ( SUnSET ) [49] . SUnSET is a non-radioactive method for monitoring protein synthesis that uses incorporation of puromycin into nascent polypeptide chains and causes termination of elongation . Puromycin-tagged polypeptides are then detected by immunoblotting with anti-puromycin antibody . In HEK 293 Trex cells expressing protein 169 , protein synthesis was inhibited increasingly from 8 h post induction ( Fig 6A ) . In contrast , in a control cell line expressing C6 . TAP inducibly [50] no such inhibition was seen ( Fig 6B ) . The effect of 169 on protein synthesis during VACV infection was tested next . HeLa cells were infected with v169 or vΔ169 , and puromycin was added at different times p . i . ( Fig 6C ) . Host protein synthesis was inhibited by 6 h p . i . and more profoundly thereafter , but no difference was detected between v169 and vΔ169 . This could be due to both viruses expressing the de-capping enzymes D9 and D10 that have profound effects on virus protein synthesis [25 , 26 , 32] and might mask effects of protein 169 . This result is consistent with the observations that protein 169 is absent from virus factories ( Figs 1D and 2 ) , and does not affect virus replication and spread ( Fig 3 ) , suggesting that protein 169 might preferentially target host protein synthesis . To determine at which stage of protein synthesis protein 169 might be acting , polysomes were profiled in HEK 293 Trex 169 cells with or without protein 169 expression ( Fig 7A and 7B ) . Cytoplasmic extracts were prepared in the presence of cycloheximide to retain intact monosomes and polysomes and these were analyzed by sucrose density gradient centrifugation . The RNA and protein composition of the gradient was measured by absorbance ( A254 nm ) and immunoblotting , respectively . Protein 169 expression caused an increase in 80S ribosomes and decrease in polysomes ( Fig 7B ) , indicating an inhibition of translational initiation . Immunoblotting of gradient fractions revealed that protein 169 co-purified partially with the 40S ribosomal fraction ( Fig 7B ) , consistent with immunofluorescence data ( Fig 2A ) . For comparison , HEK 293 Trex C6 . TAP cells were analyzed in parallel and protein C6 expression caused no such alterations to polysomes or 80S monosomes ( Fig 8A and 8B ) . To investigate whether the 80S ribosomes accumulating in the presence of protein 169 contain mRNA , polysome profiling was repeated in a higher salt buffer ( 400 mM KCl ) , conditions in which 80S ribosomes lacking mRNA dissociate into constituent subunits . However , in the presence of protein 169 , the 80S peak remained stable in high salt ( Fig 9B ) , indicating that the 80S ribosomes are associated with mRNA . Increasing the concentration of salt in the sucrose density gradient reduced the sharpness of the peaks obtained . To confirm that this effect was due to the high salt concentration , the polysome profile of uninduced HEK 293 Trex 169 cytoplasmic cell lysates was examined on sucrose gradients ( Fig 9C and 9D ) . Again , the high salt condition affects the overall sharpness of polysomal fractions independently of the expression of protein 169 . Since protein 169 inhibited cap-dependent and FMDV IRES-dependent translation , both of which require the concerted action of multiple eIFs , we tested whether protein 169 could affect translation from the cricket paralysis virus ( CrPV ) intergenic region ( IGR ) IRES . This IRES uses an unusual mechanism of translation initiation binding directly to the 40S subunit and initiating from the A-site without the requirement for initiation factors [51] . A plasmid encoding a bicistronic gene in which renilla luciferase is translated in a cap-dependent manner and firefly luciferase is translated in a CrPV IRES-dependent manner was transfected into HEK 293T cells together with A49 . TAP , 169 or empty vector control . Luciferase levels were measured by luminescence ( Fig 10A and 10B ) and protein expression was determined by immunoblotting ( Fig 10C ) . Low levels of both firefly and renilla luciferase were found in cycloheximide-treated cells as well as in cells expressing protein 169 indicating that protein 169 can inhibit translation from an IRES that does not require the activity of any initiation factors . Taken together these data indicate that protein 169 inhibits the initiation of translation causing accumulation of 80S ribosomes and that this applies to both cap-dependent and IRES-dependent translation . This generic shut-down of host protein synthesis , while virus protein synthesis remains largely unaffected , affects the expression of many proteins induced by activation of innate immune sensing pathways and results in inhibition of innate immunity within infected cells . Such a strategy would be predicted to affect the outcome of infection in vivo and therefore this hypothesis was investigated . The contribution of 169 to virus virulence was examined using two murine models of infection . The intranasal ( i . n . ) model represents a systemic infection , where the virus replicates in the lungs and spreads to other organs . Virus virulence is assessed by measuring weight loss , virus titers and signs of illness [52 , 53] . In the intradermal ( i . d . ) model , mice are inoculated by intradermal injection into the ear pinna , which results in a localized infection , and virulence is determined by measuring lesion size and healing time [54 , 55] . In the i . n . model , infection with vΔ169 resulted in significantly greater weight loss from day 5 onwards and more severe signs of illness than control viruses ( Fig 11A ) . To investigate the basis for these differences , the levels of cytokines and chemokines in broncho-alveolar lavage ( BAL ) fluids were measured early ( 24 h ) p . i . This showed that there were enhanced levels of IL-2 , IL-6 , TNF-α , CCL11 , CXCL9 and CXCL10 following infection with vΔ169 compared to both control viruses , whereas the levels of CCL2 , CCL9 , IL-12 and IL-15 were unchanged ( Fig 11B and 11C ) . Furthermore , infection with vΔ169 caused increased lung weights and the number of cells in BAL fluids on days 4 and 7 p . i . compared to control viruses ( Fig 11D and 11E ) . Measurement of lung virus titers showed that all three viruses had replicated to the same extent on days 2 and 4 p . i . , but by day 7 the titer of vΔ169 had decreased more than controls , indicating more rapid clearance ( Fig 11F ) . These observations show that infection with vΔ169 caused a greater inflammatory response , with elevated synthesis of several cytokines and chemokines , enhanced recruitment of cells into BAL fluids and more rapid virus clearance . To analyze the nature of cells recruited into BAL fluids , the cells were stained with monoclonal antibodies and quantified by flow cytometry . The majority of inflammatory cells recruited during infection were macrophages ( Fig 12A ) and lymphocytes ( Fig 12C ) including CD4+ and CD8+ T-cells ( Fig 12F , 12G and 12H ) , with fewer numbers of neutrophils , ( Fig 12B ) , NK cells ( Fig 12D ) and B cells ( Fig 12E ) . Notably on days 4 and 7 p . i . the recruitment of macrophages , total lymphocytes , T-cells , and CD4+ and CD8+ T-cells was increased following infection with vΔ169 compared to controls , and these differences may explain the more rapid clearance of this virus . In contrast , neutrophils ( Fig 12B ) , NK cells ( Fig 12D ) , and B cells ( Fig 12E ) showed no difference between all viruses . Changes in the inflammatory response to primary infection can alter the adaptive response and subsequent protection against virus challenge . This has been observed with VACV mutants that either have increased virulence , such the VACV WR strain lacking the soluble chemokine binding protein A41 [56–58] , or decreased virulence , such as the inhibitor of IRF-3 activation C6 [45 , 59 , 60] and the inhibitor of apoptosis and NF-κB activation N1 [47 , 61 , 62] . A more severe primary infection can also lead to better protection [57] , and to test whether enhanced immune response generated by vΔ169 is advantageous and would lead to better protection , the potency of vΔ169 as a vaccine was evaluated . Mice were immunized via the i . n . route with v169 , vΔ169 or v169-rev and then were challenged with wild type virus i . n . at day 28 ( Fig 13A ) . In this model , vΔ169 induced better protection against challenge as shown by reduced weight loss compared to controls ( Fig 13A ) . To investigate the basis for this , the levels of VACV neutralizing antibodies were determined by plaque reduction neutralization assay ( Fig 13B ) and the cytotoxicity of NK cells on uninfected YAC-1 cells ( Fig 13C ) and CD8+ splenic T-lymphocytes ( Fig 13D ) on VACV-infected P815 cells was measured by chromium release assay . At day 28 p . i . all groups of immunized mice had high serum antibody titers that did not differ between the groups ( Fig 13B ) . Similarly , the cytotoxicity of splenic NK cells on YAC-1 cell targets did not differ between the groups ( Fig 13C ) . However , the lysis of target cells by splenic CD8+ T-cells within the total splenocyte population from mice infected with vΔ169 was significantly greater than lysis by cells from mice infected by control viruses ( Fig 13D ) . Collectively , these data show that immunization with vΔ169 generates stronger CD8+ T-cell immunological memory and better protection against challenge . The virulence and immunogenicity of vΔ169 was also assessed after intradermal ( i . d . ) infection ( Fig 14 ) . vΔ169 caused a statistically significant increase in lesion size and duration compared to control viruses ( Fig 14A ) . Furthermore , as observed in the i . n . model , viral titers in the ears showed that all viruses replicated to a similar extent initially ( day 3 and 6 ) , but thereafter ( days 10 and 14 ) viral titers were lower for vΔ169 compared to controls ( Fig 14B ) . Additionally , mice immunized via the i . d . route with v169 , vΔ169 or v169-rev were challenged with wild type virus i . n . at day 28 ( Fig 14C ) . As observed for i . n . model , vΔ169 induced better protection against challenge as shown by reduced weight loss of mice immunized with vΔ169 compared to controls .
A functional study of VACV WR protein 169 is presented . This small , highly charged protein is expressed early during VACV infection , localizes in cytoplasmic puncta but is excluded from virus factories , and inhibits the initiation of cap-dependent and cap-independent protein synthesis . Thereby , protein 169 reduces production of host inflammatory mediators induced by activation of multiple innate immune signaling pathways . Protein 169 is conserved in many VACV strains and orthopoxviruses but nonetheless is non-essential for virus replication or spread in tissue culture . Instead , it affects the outcome of infection in vivo by decreasing the recruitment of inflammatory leukocytes , delaying clearance of virus , reducing the memory CD8+ T-cell response and diminishing protection against subsequent virus challenge . The ability of protein 169 to inhibit the innate immune response , while not affecting virus replication in cell culture or in vivo , is characteristic of many VACV immunevasins [8] . However , a striking difference between many of the immunevasins characterized hitherto and protein 169 is that the former are often inhibitors of an individual innate immune signaling pathway ( or sometimes two pathways ) by binding to one or two specific host proteins . In contrast , protein 169 is a general inhibitor of protein synthesis and targets multiple pathways that require nascent protein synthesis . Thus , by blocking the translation of host mRNAs that are transcribed , for instance , following activation of NF-κB , IRF-3 or the JAK/STAT signaling pathways , there is a decreased production of many inflammatory mediators and consequential reduced recruitment of leukocytes to the site of infection . The inhibition of host protein synthesis by viruses is widespread , but hitherto has been considered largely a strategy by which viruses subvert host metabolism to increase virus protein synthesis and production of virions . VACV protein 169 illustrates another purpose , namely , the decrease of host protein synthesis without a concomitant increase in production of virus proteins or infectious virus particles , but with the consequence of restricting the host innate immune response to infection , so aiding virus escape and diminishing immunological memory . Protein 169 is well adapted to this purpose for it is excluded from virus factories , the site of virus protein synthesis , and so targets host translation preferentially , and its loss does not affect virus replication in cell culture or in vivo . Protein 169 is also unusual in that it targets both cap-dependent and cap-independent translation . Many RNA viruses exploit IRES-dependent translation to manufacture their proteins while disabling cap-dependent translation of host mRNAs by targeting the eIF4F complex . Popular strategies are ( i ) cleavage of eIF4G by viral proteases [63–65] , ( ii ) cleavage of poly A-binding protein [66 , 67] , and ( iii ) decreasing phosphorylation of cap-binding protein eIF4E [68 , 69] . In contrast , DNA viruses use mostly cap-dependent translation and stimulate eIF4F formation . Herpes simplex virus type 1 ( HSV-1 ) protein ICP0 promotes phosphorylation of eIF4E and 4E-binding protein 1 ( 4E-BP1 ) that leads to degradation of 4E-BP1 and stimulation of formation of the eIF4F complex [70] . Also , HSV-1 protein ICP6 binds eIF4G to enhance eIF4F assembly [71] . VACV stimulates eIF4F complex formation through hyper-phosphorylation of 4E-BP1 enabling interaction between eIF4E and eIF4G [27] . Protein 169 acts differently , but has some similarity with the modulation of protein synthesis by hepatitis C virus ( HCV ) in that it leads to alterations in innate immunity . HCV relies mainly on IRES-dependent translation and causes stimulation of protein kinase R that leads to translation inhibition through phosphorylation of eIF2α to inhibit production of IFN stimulated genes ( ISGs ) [72] . However , the factors responsible for these changes and their mechanism of action remain unknown . Protein 169 is the third VACV polypeptide shown to inhibit protein synthesis , the others being the de-capping enzymes D9 and D10 [25 , 32 , 33] . These enzymes are made either early or late during infection and de-cap both host and viral mRNAs , although some preferential affinity for different cap structures have been shown [33] . Since the viral mRNAs are synthesized in greater abundance , these soon become predominant and so virus proteins are made while host protein synthesis declines . Rapid mRNA turnover is also important for progression between early , intermediate and late stages of VACV gene expression . The importance of de-capping for virus replication is illustrated by the loss or mutation of protein D10 that results in a smaller plaque phenotype , accumulation of early transcripts , lower virus yield [26] and attenuation in vivo [34] . Recently a VACV strain expressing catalytically dead versions of D9 and D10 was shown to induce large amounts of dsRNA . This activates pathways leading to inhibition of protein synthesis and consequently reduces virus production and results in severe attenuation in vivo [35] . In contrast , loss of protein 169 has no effect on virus replication in vitro ( Fig 3 ) or in vivo ( Figs 11 and 14 ) and its loss causes an increase in virulence in both i . n . and i . d . models of infection ( Figs 11 and 14 ) . In the i . n . model , infection by vΔ169 caused enhanced production of several cytokines ( IL-2 , IL-6 and TNF-α ) and chemokines ( CCL11 , CXCL9 and CXCL10 ) within 1 day p . i . ( Fig 11 ) and subsequent greater recruitment of macrophages and CD4+ and CD8+ T cells ( Fig 12 ) and increased lung weight ( Fig 11D ) . Later , this greater recruitment of inflammatory cells leads to more rapid virus clearance and recovery ( Fig 11 ) . Similarly , in the i . d . model the greater inflammatory response is reflected in a greater lesion size , but again this is followed by more rapid virus clearance and recovery ( Fig 14 ) . The early expression of protein 169 is consistent with prior RNA analysis of the VACV genome that showed early transcription of this ORF and 169 mRNAs were detected from 1 h p . i . [73 , 74] . Sometimes viruses that induce exacerbated immune responses are more virulent and in this regard it is notable that orthopoxviruses lacking ORF 169 are generally of high virulence . For instance , all sequenced variola viruses and ectromelia virus lack ORF 169 and these viruses are highly virulent in man or mice , causing smallpox and mousepox , respectively . Similarly , VACV strain Copenhagen lacks ORF 169 and caused a higher frequency of post-vaccination complications in man than the more widely used VACV strains Lister and New York City Board of Heath ( Wyeth ) [1] . VACV strain Copenhagen also caused larger lesion sizes in the mouse intradermal model in comparison to other VACV strains used as smallpox vaccines in man [55] . However , VACV strain Copenhagen and all variola virus strains also lack another factor that diminishes virulence , namely the soluble IL-1β binding protein encoded by gene B15R of VACV strain WR [53 , 75] , and the causes of enhanced virulence are probably multi-factorial . The increased virulence seen by loss of gene 169R has a few parallels in orthopoxvirus biology . In addition to deletion of the soluble IL-1β receptor encoded by VACV WR mentioned above [53] , deletion of the chemokine binding protein A41 [56] , and the B13 serine protease inhibitor [55] each caused an increase in virulence in either the i . n . or i . d . model , and in some cases also induced a stronger immunological memory response that resulted in better protection against virus challenge [57 , 76] . Infection with vΔ169 generated a stronger innate response ( Figs 11 and 12 ) , that led to a stronger memory CD8+ T cell response and better protection to virus challenge ( Figs 13 and 14 ) . Increased immunological memory responses and better protection against challenge have also been observed with VACV mutants with diminished virulence , such as viruses lacking the C6 or N1 proteins [60 , 77] . Protein 169 localizes mainly in the cytoplasm of infected cells throughout the course of infection . The punctate pattern observed might suggest co-localization of 169 with some specific organelles , but only some partial overlap with 40S ribosomes was observed . The precise mechanism by which protein 169 inhibits translation remains to be determined , but the polysome profiling experiments described ( Figs 7–9 ) reveal that protein 169 expression leads to an accumulation of 80S monosomes and reduction of polysomes , particularly of heavier polysomes . This pattern is consistent with a reduced rate of translation initiation , and the stability of the 80S monosomes in high-salt indicates that the 80S ribosomes are mRNA-associated , rather than present in a free pool [78] . Reducing a pool of free ribosome is a strategy used by cardiovirus protein 2A that , in contrast to protein 169 , causes accumulation of monosomes free of mRNA [79] . A direct interaction between protein 169 and either the mRNA cap or 40S subunit was not observed , nor was an effect of protein 169 on translation in vitro using rabbit reticulocyte lysate . However , we cannot be sure whether the prepared fraction of protein 169 is functional under the conditions tested . Nonetheless , the capacity of protein 169 to block FMDV IRES-directed translation initiation is consistent with an eIF4E-independent inhibitory mechanism . In addition , the inhibition of CrPV IRES-dependent translation by protein 169 suggests that its inhibitory activity is not mediated by interference with other eIFs . In summary , 169 is an inhibitor of cap-dependent and cap-independent translation , it affects virus virulence and contributes to VACV immunogenicity by diminishing the innate and adaptive immune response . This study illustrates that viral inhibition of protein synthesis can be an immune evasion strategy rather than a mechanism to increase yields of virus from infected cells .
This work was carried out in accordance with regulations of The Animals ( Scientific Procedures ) Act 1986 . All procedures were approved by the United Kingdom Home Office and carried out under the Home Office project licence PPL 70/7116 . BSC-1 ( ATCC CCL-26 ) , CV-1 ( ATCC CCL70 ) , HEK 293T ( ATCC CRL-11268 ) and A549 ( ATCC CCL-185 ) cells were maintained in Dulbecco’s modified minimal essential medium ( DMEM ) containing 10% fetal bovine serum ( FBS ) and penicillin/streptomycin ( 50 μg/ml ) . RK-13 ( ATCC CCL-37 ) and TK-143 ( ATCC CRL-8303 ) cells were grown in minimum essential medium ( MEM ) and supplemented as above . HeLa ( ATCC CCL-2 ) cells were grown in MEM with addition of non-essential amino acids ( 1% ) and supplemented as above . HEK 293 Trex ( Invitrogen ) cells were maintained in DMEM containing 10 μg/ml blasticidin , 100 μg/ml zeocin and supplemented as above . The sequence of the VACV WR 169R gene was codon optimized by GENEART for expression in mammalian cells . 169R was then sub-cloned into mammalian expression vectors pcDNA 3 . 1 or pcDNA4 TO ( Invitrogen ) without a tag or with an N-terminal FLAG tag . E . coli expression plasmid pOPINE were engineered to express a 169R wild type sequence with a C-terminal His tag ( 169-His ) and plasmid pGEX-6p-1 was engineered to express a 169R wild type sequence with an N-terminal glutathione S-transferase ( GST ) tag ( GST-169 ) . Plasmid Z11-Δ169 was used to construct the VACV mutant lacking gene 169R and contained flanking regions of the 169R gene locus cloned into plasmid Z11 that contains the E . coli guanine phosphoribosyltransferase ( Ecogpt ) fused with enhanced green fluorescent protein ( EGFP ) driven by an early/late VACV promoter as described [45] . Plasmid Z11-169-rev was used to construct the revertant virus v169-rev and contains the 169R gene and flanking sequences inserted into Z11 plasmid . A plasmid encoding a bicistronic gene expressing firefly luciferase in a cap-dependent manner and renilla luciferase in a FMDV IRES-dependant manner was a kind gift from Prof . Ian Goodfellow , Department of Pathology , University of Cambridge . A plasmid encoding a bicistronic reporter gene expressing firefly luciferase in a cricket paralysis virus ( CrPV ) IRES-dependent manner and renilla luciferase in a cap-dependant manner was a kind gift from Dr . Eric Jan , Department of Biochemistry and Molecular Biology , University of British Columbia , Canada . NF-κB-Luc , ISRE-Luc and TK renilla was obtained from Dr . Andrew Bowie ( Trinity College , Dublin , Ireland ) , ISG56 . 1 Luc was from Ganeth Sen ( Lerner Research Institute , Ohio ) , and M5P Luciferase–NEMO ( Luc-NEMO ) and M5P GFP-FLAG were obtained from Dr . Felix Randow ( MRC Laboratory of Molecular Biology , Cambridge , United Kingdom ) . C6 . TAP , N1 . TAP , B14 . FLAG and A49 . TAP were described previously [15 , 17 , 45 , 47] . V5-PiV5-V was provided by Jennifer H . Stuart ( Department of Pathology , University of Cambridge , UK ) . Rabbit polyclonal antiserum raised against recombinant 169 protein was used for immunoblotting ( diluted 1:1000–2000 ) and purified anti-169 770 P antibody was used for immunofluorescence ( diluted 1:50 ) . Other antibodies used were mouse anti-FLAG ( Sigma , F1804 , diluted 1:1000 ) , rabbit anti-FLAG ( Sigma-Aldrich , F7425 , diluted 1:5000 ) , anti-D8 mouse mAb AB1 . 1 against VACV protein D8 [39] ( diluted 1:500 ) , anti-α-tubulin ( Millipore , 05–829 , diluted 1:5000 ) , anti-actin ( Sigma , A2066 , diluted 1:1000 ) , anti-lamins A+C ( Abcam , ab898 , diluted 1:1000 ) , anti-VACV protein C16 [38] ( diluted 1:1000 ) , anti-protein disulphide isomerase ( PDI , 1D3 clone , Enzo Life Sciences , diluted 1:50 ) , anti-GM130 ( Transduction laboratories , diluted 1:300 ) , anti-clathrin ( Abcam , diluted 1:50 ) , anti-human transferrin receptor ( Zymed , used at 2 . 5 μg/ml ) , anti-ribosomal protein S6 ( Cell Signalling , diluted 1:25 for immunofluorescence and 1:1000 for immunoblotting ) , anti-eIF4AI ( Santa Cruz Biotechnology , N-19 , diluted 1:500 dilution , a kind gift from Prof . Ian Goodfellow ) , anti-eIF4E ( Santa Cruz Biotechnology , A-10 diluted 1:500 ) , anti-ribosomal protein L29 ( Santa Cruz Biotechnology , P-14 , dilution 1:200 ) , anti-puromycin ( Millipore , clone 12D10 , diluted 1:15000–1:25000 ) , Alexa Fluor 488 goat anti-rabbit IgG ( H+L ) ( Invitrogen , A-11008 , diluted 1:750 ) , Alexa Fluor 549 donkey anti-mouse IgG ( H+L ) ( Invitrogen , A-10036 , diluted 1:750 ) , and MitoTracker Red CM-H2XRos ( Invitrogen , M7513 , diluted 1:5000 ) . Reagents used in this study were puromycin ( InvivoGen ) , cycloheximide ( Calbiochem ) , doxycycline ( Melford ) , blasticidin ( Gibco ) and zeocin ( Invitrogen ) . VACV vΔ169 was constructed by transfecting plasmid Z11-Δ169 into VACV WR infected CV-1 cells using FuGENE 6 and a recombinant VACV was isolated by transient dominant selection [40] as described for other VACV deletion mutants [12 , 80] . Plaque purified wild type 169 ( v169 ) and deletion 169 ( vΔ169 ) viruses were isolated from the same intermediate virus and were genotyped using PCR and primers amplifying the flanking regions of the 169R locus . The revertant 169 virus ( v169-rev ) was constructed by transfection of plasmid Z11-169-rev into vΔ169-infected CV-1 cells following the same procedure as described above . Genomic DNA isolated from recombinant VACVs ( v169 , vΔ169 and v169-rev ) were compared to parental VACV WR virus using restriction endonuclease digestion with HindIII or SphI digestion and virus DNA was visualized after pulsed field gel electrophoresis . E . coli BL21 ( DE3 ) R3 pRARE cells ( kind gift from SGC Oxford ) , where R3 denotes a derivative of BL21 ( DE3 ) resistant to a strain of T1 bacteriophage ( SGC Oxford ) and the pRARE plasmid originates from the Rosetta strain ( Novagen ) and supplies tRNAs for rare codons , were transformed with the 169-His expression plasmid . The bacteria were grown in terrific broth and the expression of 169-His was induced by 1 mM IPTG at 37°C for 6 h . Bacteria were collected by centrifugation , lysed and disrupted by sonication . 169-His was purified from the soluble fraction by immobilized metal affinity chromatography ( IMAC ) using a His-Trap HP column followed by ion exchange chromatography ( IEX ) using a MonoQ GL column . Three and a half mg of 169-His was used to inoculate two rabbits ( Eurogentec , Seraing , Belgium ) to obtain polyclonal sera . Two rabbits ( number 770 , 771 ) were immunized at day 0 , 14 , 28 and 56 with Freund's complete adjuvant at day 0 and with incomplete Freund's adjuvant for the boosts with dose of 400 μg of 169-His . Sera prepared from venous blood drawn before immunization and at day 66 were tested for recognition of protein 169 expressed during VACV infection . Serum from rabbit 771 was sensitive enough to detect protein 169 from VACV-infected cells . This serum was used for immunoblotting analysis throughout this study ( further referred as anti-169 ) . Serum from rabbit 770 was further purified against GST-169 using AminoLink immobilization kit . GST-169 protein was produced in BL21 ( DE3 ) E . coli bacteria ( Merck Millipore ) transformed with pGEX-6p-1 GST-169 plasmid . Bacteria were grown in LB and expression of GST-169 was induced by 1 mM IPTG at 37°C for 6 h . Bacteria were collected by centrifugation , lysed and disrupted by sonication . GST-169 was purified from soluble fraction using glutathione-sepharose 4B and size exclusion chromatography ( SEC ) using Superdex 75 10/300 GL column . Two mg of GST-169 was used for polyclonal serum purification using AminoLink immobilization kit following the manufacturer’s instructions for the pH 7 protocol . Protein 169-specific purified IgG ( further referred as an anti-169 purified antibody ) were used for immunofluorescence studies . For analysis of virus single step growth properties , BSC-1 cells were infected at 10 PFU/cell for 12 or and 24 h . Extracellular virus in the clarified growth medium ( after centrifugation at 500 x g for 10 min ) was titrated by plaque assay on BSC-1 cells . Cell associated virus was measured by scraping cells from the plastic flask , combining these with the debri from the supernatant and collection by centrifugation as above . Cells were then disrupted by three rounds of freeze-thawing and sonication and the virus was titrated by plaque assay on BSC-1 cells . For analysis of multiple step growth properties , BSC-1 cells were infected at 0 . 05 PFU/cell for 24 and 48 h . The extracellular and cell-associated viral titers were determined as described above . BSC-1 , RK-13 and TK-143 cells were infected with the indicated VACVs at 50 PFU/ well of a 6-well plate . The radius of plaques was measured after 72 h using Axiovision 4 . 8 . 2 software on an Axiovert . A1 microscope ( Zeiss ) with Axiocam MRc . In each condition 20 plaques per virus were measured in three independent experiments . For intranasal ( i . n . ) model of infection , BALB/c mice ( 6–8 weeks old ) were inoculated with VACVs , which had been purified by sedimentation twice through a sucrose cushion , ( 5 × 103 PFU into each nostril ) and monitored daily for a weight loss and scored for signs of illness as follows hair ruffling , back arching , reduced mobility , pneumonia [52 , 53] . For the intradermal ( i . d . ) model of infection , female C57BL/6 mice ( 6–8 weeks old ) were inoculated with purified VACVs ( 104 PFU ) in both ear pinna and the diameter of the lesion was measured daily using a micrometer [54] . The administered dose was confirmed by plaque assay . For challenge experiments , immunized animals were challenged i . n . 28 d p . i . with 5 × 106 PFU of v169 and weighed daily thereafter . Bronchial alveolar lavage ( BAL ) fluids were prepared on the indicated days . These were centrifuged at 1500 g to obtain cells for flow cytometry and the clarified supernatant was used for ELISA . Live cells collected from BAL fluids were counted using a haemocytometer following staining with trypan blue . For determination of lung and ear tissues viral titers , the lungs and ears tissues were homogenized and washed through a 70 μm nylon mesh using DMEM and 10% FBS . Cells were then frozen and thawed three times , and sonicated thoroughly to liberate intracellular virus . Infectious virus was titrated in duplicate by plaque assay on BSC-1 cell monolayers . For chromium-release cytotoxicity assay , NK cell cytotoxicity and VACV-specific cytotoxic T lymphocyte ( CTL ) activity within total splenocyte populations was assayed with a standard 51Cr-release assay as described [77] . NK-mediated lysis was tested on uninfected YAC-1 cells , while VACV-infected P815 cells ( H-2d , mastocytoma ) were used as targets for VACV-specific CTL lysis . The percentage of specific 51Cr release was calculated as specific lysis = [ ( experimental release−spontaneous release ) / ( total detergent release−spontaneous release ) ]×100 . The spontaneous release values were always < 10% of total lysis . Anti-mouse CD3 ( clone 145-2C11 ) , CD4 ( GK1 . 5 ) , CD8 ( 5H10-1 ) , CD45 ( 30-F11 ) , CD45R ( RA-6B2 ) , NK1 . 1 ( PK136 ) , CD11b ( M1/70 ) , F4/80 ( BM8 ) , Ly-6G/Ly-6C ( RB6-8C5 ) , Ly6G ( 1A8 ) and CD16/32 ( 2 . 4G2 ) mAb were purchased from BD Biosciences or Biolegend . The mAbs were purified or conjugated with FITC , PerCP/cy5 . 5 , APC , PE-Cy7 , APC/Cy7 , BV650 or PE . Isotype controls were used as negative controls . Flow cytometry was performed with a BD LSR Fortessa ( BD Biosciences ) , and data were analyzed with FlowJo software ( Tree Star Inc . ) . Events were gated for live lymphocytes on foward scatter × side scatter and dead cells were excluded on the basis of atypical fluorescence . Data were further analyzed using Prism ( GraphPad , La Jolla , CA , USA ) . Cytokines ( IL-2 , IL-6 , IL-12 , IL-15 and TNF-α ) and chemokines ( CCL2 , CCL7 , CCL11 , CXCL9 and CXCL10 ) levels in the supernatants of BAL were determined following i . n . infection of 5 x 103 dose in 100 μl 24 h p . i . using DuoSet ELISA kits ( R&D Systems Inc . ) and were carried out according to the manufacturer's instructions . HeLa cells were either mock-infected or infected at 10 PFU/cell for 7 h . The cells were fractionated using Cell fractionation kit ( Thermo Scientific ) according to the manufacturer’s instruction . HeLa cells were seeded on glass coverslips and were either mock-infected of infected at 10 PFU/cell or 2 PFU/cell in case of 16 h time point . At the indicated times , the cells were washed twice with PBS and fixed with 4% paraformaldehyde in PBS containing 250 mM HEPES . The cells were permeabilized with 0 . 1% triton X-100 followed by blocking with 10% FBS in PBS ( blocking buffer ) for 0 . 5 h . Coverslips were incubated with primary antibodies for 1 h in a moist chamber followed by three 10 min washes with 10% FBS . Coverslips were incubated with secondary antibody ( Alexa Fluor 488 Goat Anti-Rabbit IgG ( H+L ) , Alexa Fluor 546 donkey Anti-Mouse IgG ( H+L ) 1:750 diluted in blocking buffer ) for 30 min in a moist chamber followed by three 5 min washes with 10% FBS and PBS only . The coverslips were washed with water and mounted in Mowiol 4–88 containing DAPI . Coverslips were allowed to set and stored at 4°C . Cells were visualized by Axio observer Z1 confocal microscope ( Zeiss ) with a 63x oil objective . Reporter gene assays was performed in HEK 293T cells in 96-well dishes as described [45] . Cells were transfected in triplicate with 60 ng of firefly reporter plasmid ( NF-κB , ISG 56 . 1 or ISRE ) , 10 ng of TK renilla ( as an internal control ) and 100 ng of expression plasmid or empty vector control using TransIT-LT1 according to the manufacturer’s instruction . The following day cells were stimulated; ( i ) with 75 ng of TNF-α for 7 h ( NF-κB Luc ) or ( ii ) transfected with 200 ng/well of poly I:C for 24 h ( ISG56 . 1 Luc ) using lipofectamine , or ( iii ) with 100 U/ml of IFN-α for 7 h ( ISRE Luc ) . Cells were lysed using passive lysis buffer ( Promega ) and firefly luciferase activity was normalized to the renilla luciferase activity , and these data were further normalized to the un-stimulated controls of each test plasmid . A549 cells were transfected in triplicate with GFP . FLAG , B14 . FLAG , C6 . TAP and 169 using Lipofectamine LTX Plus ( Life Technologies ) . The following day cells were stimulated with 50 ng/ml of TNF-α for 7 h . RNA was extracted using RNeasy Mini Kit ( Qiagen ) and converted to cDNA using SuperScript reverse transcriptase . ICAM-1 , IL-6 and NF-κBia mRNA were quantified in comparison to hypoxanthine-guanine phosphoribosyltransferase ( HPRT ) using SYBR green master mix . HeLa or HEK 293T cells were transfected with indicated plasmids for 24 h . RNA was extracted using RNeasy Mini Kit ( Qiagen ) and converted to cDNA using SuperScript reverse transcriptase . GFP , luciferase or 169 mRNA were quantified and compared to HPRT or glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) . For analysis of cytoplasmic and nuclear mRNA , HEK 293T cells were transfected with empty vector control , A49 . TAP and 169 together with NEMO-Luc . After 4 h the cells were treated with CHX ( 1 μg/ml ) for 16 h . Cells were lysed in RLN buffer ( 50 mM Tris HCl pH 8 . 0 , 140 mM NaCl , 1 . 5 mM MgCl2 , 0 . 5% ( v/v ) Nonidet P-40 , 1 mM DTT , 500 U/ml RNAse out ) , scraped and incubated for 5 min on ice . Nuclei were sedimented by centrifugation at 1000 g for 3 min . Supernatant ( cytoplasmic fraction ) was taken and mRNA was extracted according to the manufacturer’s instruction ( Qiagen ) . RLT buffer was added to the pellet ( nuclear fraction ) and forced through a 25G needle ten times . Further steps followed the manufacturer’s instructions ( Qiagen ) . cDNA was prepared using SuperScript reverse transcriptase . Luc-NEMO , HPRT and TATA box binding protein mRNA were quantified in comparison to ( GAPDH ) using SYBR green master mix . HEK 293T cells were transfected in triplicate with GFP . FLAG , B14 . FLAG , C6 . TAP , 169 and Δ12 A49 . TAP . The following day cells were either mock-infected or infected with SeV for 24 h . The amount of CXCL10 in the supernatant was determined using human CXCL10 Quantikine ELISA Kit ( R&D Systems ) . The results were analyzed using nonlinear standard curves for ELISA ( GraphPad PRISM ) . HEK 293 Trex ( Invitrogen ) empty cells were transfected with 169 ( pcDNA4 TO ) using TransIT-LT1 . Transfected cells were selected in the presence of zeocin and were serially diluted to obtain individual clones . Expression of protein 169 within these clones were analyzed by immunoblotting and immunofluorescence . In the chosen clone at least 90% of cells were expressing protein 169 . HEK 293 Trex 169 or C6 . TAP [50] cells were treated with 1 μg/ml of DOXY for the indicated times to express protein 169 or C6 . Cells were treated with 5 μg/ml of puromycin for 25 min and harvested for analysis by immunoblotting [49] . HeLa cells or BSC-1 cells were mock-infected or infected with VACVs at 5 PFU/cell for the indicated times . Cells were treated with puromycin as described above . HEK 293 Trex 169 or C6 . TAP [50] cell were uninduced or induced with 1 μg/ml of DOXY for 16 h . Thirty min prior to harvesting , the cells were treated with CHX ( 1 μg/ml ) . Cells were washed and lysed in lysis buffer supplemented with protease inhibitors ( cOmplete , Mini , EDTA-free , Roche , 1 tablet in 10 ml of lysis buffer ) ( 20 mM Tris HCl pH 7 . 5 , 100 mM KCl , 5 mM MgCl2 , 1 mM CHX , 1 mM DTT , 0 . 1 mM EDTA ) with DNAse I and NP-40 ( 0 . 03% ) followed by trituration with a 25G needle . Cleared ( 19 , 000 g for 5 min at 4°C ) cytoplasmic lysates were layered on top of sucrose density gradient ( 10–50% sucrose in lysis buffer ) prepared by a Gradient Master ( Biocomp ) and resolved by centrifugation at 200 , 000 g for 90 min at 4°C . Absorbance ( 254 nm ) composition within the gradient was measured during fractionation at 4°C using an Isco fractionator . Proteins from these fractions were extracted using methanol-chloroform extraction and subjected to immunoblotting analysis . Polysome profiling in higher salt condition was carried out with HEK 293 Trex 169 as described above except that the lysis buffer and sucrose density gradient contained 400 mM KCl . Statistical analysis was performed using Student’s two tail t-test unless otherwise stated . | Long after smallpox was eradicated by vaccination with vaccinia virus , the study of this virus continues to reveal novel aspects of the interactions between a virus and the host in which it replicates . In this work we investigated the function of a previously uncharacterized VACV protein , called 169 . The results show that protein 169 inhibits the synthesis of host proteins in cells and thereby provides a broad inhibition of the host innate immune response to infection . Unlike several other virus inhibitors of host protein synthesis , protein 169 acts by inhibiting the initiation of protein synthesis by both cap-dependent and cap-independent pathways . Also unlike several other virus protein synthesis inhibitors , the loss of protein 169 does not affect virus replication or spread , but the virus virulence was increased . This more severe infection is , however , cleared more rapidly and results in a stronger immunological memory response that is mediated by T-cells and provides better protection against re-infection . This work illustrates how shutting down host protein synthesis can be a strategy to block the host immune response to infection rather than a means to manufacture more virus particles . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Inhibition of Translation Initiation by Protein 169: A Vaccinia Virus Strategy to Suppress Innate and Adaptive Immunity and Alter Virus Virulence |
Inactivation of the Rb tumor suppressor can lead to increased cell proliferation or cell death depending on specific cellular context . Therefore , identification of the interacting pathways that modulate the effect of Rb loss will provide novel insights into the roles of Rb in cancer development and promote new therapeutic strategies . Here , we identify a novel synthetic lethal interaction between Rb inactivation and deregulated Wg/Wnt signaling through unbiased genetic screens . We show that a weak allele of axin , which deregulates Wg signaling and increases cell proliferation without obvious effects on cell fate specification , significantly alters metabolic gene expression , causes hypersensitivity to metabolic stress induced by fasting , and induces synergistic apoptosis with mutation of fly Rb ortholog , rbf . Furthermore , hyperactivation of Wg signaling by other components of the Wg pathway also induces synergistic apoptosis with rbf . We show that hyperactivated Wg signaling significantly increases TORC1 activity and induces excessive energy stress with rbf mutation . Inhibition of TORC1 activity significantly suppressed synergistic cell death induced by hyperactivated Wg signaling and rbf inactivation , which is correlated with decreased energy stress and decreased induction of apoptotic regulator expression . Finally the synthetic lethality between Rb and deregulated Wnt signaling is conserved in mammalian cells and that inactivation of Rb and APC induces synergistic cell death through a similar mechanism . These results suggest that elevated TORC1 activity and metabolic stress underpin the evolutionarily conserved synthetic lethal interaction between hyperactivated Wnt signaling and inactivated Rb tumor suppressor .
The Retinoblastoma protein Rb is a tumor suppressor inactivated in a broad spectrum of cancers [1] , [2] . Rb functions mainly through binding to the E2F family of transcription factors and regulating the expression of diverse cellular targets involved in cell cycle regulation , DNA replication and repair , apoptosis , metabolism , as well as differentiation . Consistent with this , loss of Rb can lead to increased cell proliferation or increased cell death , depending on specific cellular contexts . Therefore identification and characterization of the genes or signaling pathways that can modulate the consequences of Rb loss in cell proliferation or cell death will significantly advance our understanding of the role of Rb in cancer development , and may potentially help the development of novel approaches for therapeutic interventions [3] . The function of Rb and E2F proteins are highly conserved and much simpler in Drosophila . These features , in conjunction with a plethora of sophisticated genetic tools , make Drosophila an ideal model to identify genes that modulates the consequences of Rb loss [4] , [5] . Forward genetic screens have identified several genes that show synergistic effects on apoptosis or differentiation with rbf ( fly Rb ) mutation [6] , [7] , [8] , [9] , [10] . Of particular interest is the synthetic lethal interactions between rbf and TSC genes [10] , [11] , which is conserved in mammalian systems [10] , [12] . TSC2 functions in a complex with TSC1 to inhibit TORC1 activity by promoting Rheb in the inactive GDP-bound form [13] , [14] . Mutations of TSC induce hyperactive TORC1 activity , which leads to excessive cellular stress , including ROS and energetic stress , and causes synergistic cell death in conjunction with Rb inactivation [9] , [10] , [12] . Consistent with this , several recent studies demonstrate that Rb also plays important roles in cell metabolism and stress induction . In Drosophila , rbf mutation was shown to cause metabolic reprogramming and rbf mutants are sensitized to conditions that impose metabolic stress such as fasting , which can be rescued by glutamine supply [15] . In C . elegans , transcriptome analysis of wild type and Rb mutant under normal or starving conditions revealed that Rb is essential not only to repress stress-inducible and metabolic genes , but also to activate stress-resistant genes , mitochondrial genes , and potential insulin pathway antagonists [16] . Furthermore , studies using mouse embryonic fibroblasts ( MEFs ) from triple knock-outs of all three Rb family members show that Rb/E2F directly regulate genes involved in glutamine metabolism [17] . Taken together , these studies suggest that Rb has conserved functions modulating cellular metabolism as well as the sensitivity of cells to additional metabolic stresses induced by specific environmental or genetic conditions . In the current study , we identify a novel synthetic lethal interaction between deregulated Wg signaling and rbf mutation through genetic screens in Drosophila . We show that mutation of axin ( axn ) , a negative regulator of the Wg signaling , significantly alters the expression of metabolic genes and is hypersensitive to metabolic stress induced by fasting , which can be rescued by glutamine supply . We further demonstrate that deregulated Wg signaling increased TORC1 activity , which induced excessive metabolic stress and synergistic cell death with rbf mutation . Finally we show that inactivation of APC and Rb induces synergistic apoptosis in human cancer cells through a similar mechanism . These results provide an alternative explanation for the long standing but confusing observation that colorectal cancers , which have deregulated Wnt signals , generally preserve Rb function and may even have amplification of the Rb loci .
In a genetic screen to identify mutations that can modulate rbf mutant phenotypes , we identified an EMS mutant 127 . In Drosophila adult eyes with mosaic clones , mutant clones are in white color and wild type cells in red color ( Fig . 1A ) . Comparing to wild-type control clones , rbf single mutant clones were generally a bit smaller while 127 single mutant clones were similar to or moderately larger than WT clones ( Fig . 1B–C ) . However , rbf and 127 double mutant clones were very small or undetectable in the adult eyes ( Fig . 1D ) , suggesting that rbf and 127 mutations have synergistic effects against clonal growth or survival . We tested whether the decreased amount of rbf and 127 ( rbf 127 ) double mutant clones in adult eyes correlated with increased apoptosis in larval eye discs . Apoptosis in eye discs can be detected by the anti-cleaved caspase3 ( C3 ) antibody . As shown previously [10] , [18] , [19] , rbf mutation caused increased apoptosis just anterior to the morphogenetic furrow ( MF ) while little apoptosis was detected in wild type cells ( GFP positive ) at this stage ( Fig . 1E ) . Although 127 mutant clones showed little apoptosis ( Fig . 1F ) , rbf and 127 double clones located anterior to the MF exhibited significantly increased level of apoptosis compared to the single mutant clones ( Fig . 1G , the results were quantified in 1N ) . The 127 mutation was mapped to the Drosophila genomic region between 99D1-99E1 where the axn gene is located . Several evidences demonstrate that 127 mutation is an allele of axn: 1 ) 127 mutation failed to complement with the previously generated axn alleles; 2 ) DNA sequencing and mRNA RACE of the axn gene in 127 mutants revealed that Exon 10 of axn is linked to a repetitive heterochromatin sequence instead of Exon 11 . Therefore the axn gene in 127 mutants encodes a protein lacking part of the DIX domain at the C-terminus ( Supplementary Fig . S1A–D ) ; 3 ) 127 homozygous mutants die at the pupal stage . Expression of wild-type Axn protein by hs-Gal4/UAS-Axn can partially rescue the pupal lethality , resulting in the development of adult flies without obvious defects; and 4 ) 127 mutant significantly increased Armadillo ( Arm , fly β-catenin ) protein levels ( Supplementary Fig . S1E ) . Therefore , we renamed 127 as axn127 . Since the phenotypes of axn127 in lethality and in cell fate changes are much weaker than the previously reported axn alleles ( see below ) , we consider axn127 as a weak axn mutant allele . To determine whether the axn mutation mediates the observed synergistic apoptosis phenotype with rbf , we tested the effects of the previously reported strong axn alleles , including axnEY10228 ( axnEY ) , axnE77 , and axnS044230 ( axnS ) [20] , [21] , [22] . Low level of apoptosis was observed in single mutant clones of these strong axn alleles , and much stronger apoptosis was observed in axn , rbf double mutant clones ( Fig . 1H–M , results were quantified in 1O ) . Consistent with the notion that axn127 is a weak allele , apoptosis in rbf- axnEY , axnE77 , or axnS double mutant clones were observed in both anterior as well as posterior of the eye discs , while apoptosis in rbf axn127 mutant cells were restricted to the region anterior to the MF . The different patterns of apoptosis in eye discs are likely due to the different effects of the strong and the weak axn alleles on cell fate determination . Photoreceptor differentiation in eye disc can be detected by staining with the neuronal marker Elav . While the strong axn alleles blocked photoreceptor differentiation ( Supplementary Fig . S2D ) [21] , [23] , axn127 did not ( Supplementary Fig . S2B ) . In addition , rbf mutation did not have obvious effects on photoreceptor differentiation either alone or together with the axn alleles ( Supplementary Fig . S2 A , C , E ) . To further compare the effects of the axn alleles on differentiation , we examined the effect of axn mutation on Senseless ( Sens ) expression , which is expressed in the SOPs along the presumptive wing margin [24] . We found that the strong axnEY mutation caused ectopic expression of Sens in wing discs , while axn127 mutation did not ( Supplementary Fig . S2 G , I ) . Again rbf mutation did not affect Sens expression either alone or together with the axn alleles ( Supplementary Fig . S2 F–J ) . Taken together , these data show that axn127 does not affect photoreceptor differentiation in contrast to the previously identified strong axn alleles , and that rbf mutation has synergistic effects with axn on apoptosis but not on cell fate determinations . To determine whether deregulated Wg signaling mediates the synergistic cell death effect of axn with rbf , we examined the effect of inactivating APC genes , which encode proteins that are in a complex with Axin protein to regulate β-catenin degradation and Wg signaling activity . As shown in Fig . 2 , Drosophila APC1-APC2 mutations also induce strong synergistic apoptosis with rbf mutation in eye discs ( Fig . 2A–B , quantified in 2K ) . Therefore deregulation of the Wg signaling by inactivation of APC also induces synergistic apoptosis with rbf mutation . We further tested the effect of deregulating Wg signaling by using dominant negative GSK3 ( GSK-DN ) or dominant active Armadillo ( Arm-DA ) . Specifically , heat shock FLP-out approach was used to express GSK-DN or Arm-DA with or without rbf-RNAi in the whole eye discs at early L3 larval stage when photoreceptor differentiation has initiated in the posterior eye disc ( Fig . 2F–J , samples with GFP shown in Fig . S1F ) . With this approach , rbf-RNAi induced a stripe of apoptosis just anterior to MF ( Fig . 2F ) , while expression of GSK-DN or Arm-DA alone did not induce obvious apoptosis ( Fig . 2G , 2I ) . However , GSK-DN or Arm-DA together with rbf-RNAi induced apoptosis in a broad region anterior to MF ( Fig . 2H , 2J , quantified in 2L ) , which is similar to the apoptosis induced by axn127 rbf mutations . Therefore , deregulation of Wg signaling using dominant-negative GSK3 or dominant-active Armadillo also induce synergistic cell death in conjunction with rbf inactivation . Furthermore , inhibiting Wg signaling by expressing dominant negative TCF ( TCF-DN ) significantly inhibited synergistic cell death observed in axnS rbf double mutant clones ( Fig . 2C–E , 2M ) , indicating that synergistic cell death of axn and rbf double mutants depends , at least in part , on the transcriptional activities of Arm/TCF . Synergistic apoptosis was also observed in axn rbf mutant clones in wing discs ( Supplementary Fig . S2K–N ) , although the apoptotic levels were significantly lower than those observed in eye discs ( Fig . 1K , 1O ) . This difference is likely associated with the different effects of axn mutation on differentiation in the wing and eye discs . As discussed above , the strong axn mutations promote wing margin SOP cell fate as shown by ectopic Sens expression ( Fig . S2I–J ) while suppress photoreceptor differentiation in eye discs as shown by the block of Elav expression ( Fig . S2D–E ) . The EGFR pathway is an important survival signal that is coupled with photoreceptor and SOP differentiation [25] , [26] , [27] , [28] . We found that EGFR signaling activities , reflected by pERK antibody staining and Argos-lacZ reporter expression , are downregulated in and posterior to the MF in eye discs ( Fig . 2N–O′ ) but are upregulated in wing discs in axn mutant clones ( Supplementary Fig . S2O–P′ ) . Therefore the differential effects of axn mutation on EGFR signaling in eye and wing tissues likely influenced the level of axn rbf synergistic cell death . In summary , these results show that hyperactivation of the Wg signaling in conjunction with rbf mutation induce synergistic apoptosis in developing imaginal discs , and that the level of apoptosis is also influenced by tissue-specific effects of Wg signaling on cell differentiation and survival signaling . We determined the effect of axn mutation on cell growth by comparing the ratio of individual mutant clone area over the corresponding WT twin spot area . Although axn127 and other strong axn mutations have different effects on cell fate determination , all axn mutant clones show increased clone growth compared to WT controls ( Supplementary Fig . S3 A–D ) . One important growth and proliferation regulator in fly imaginal discs is the TSC-Rheb-TOR pathway . To determine whether TOR signaling is affected by axn mutation , we examined the phosphorylation of S6K , a direct target of TORC1 , in eye discs that consist of mostly axn or tsc1 mutant clones . The level of phospho S6K was significantly increased in axn mutant eye discs and similar to that of the tsc1 mutant discs , which was used as a positive control ( Fig . 3A ) . Therefore TORC1 signaling activity is significantly increased by mutation of axn . A previous study showed that deregulated TORC1 increased dE2F1 protein level and promote S phase entry [11] . Indeed , increased expression of PCNA-GFP , an E2F reporter , was observed in both the strong and the weak axn mutant clones ( Supplementary Fig . S3 E–F′ ) . In addition , increased dE2F1 protein and increased BrdU incorporation were also observed in axn mutant clones ( Supplementary Fig . S3G–H′ ) . Since previous studies showed that high TORC1 activities induced synergistic apoptosis with rbf mutation [10] , [11] , we tested whether increased TORC1 signaling activity contributes to synergistic cell death in axn rbf double mutant cells . Inhibition of TORC1 activity by mutation of Rheb , a direct upstream activator of TORC1 , significantly decreased apoptosis in axnS rbf mutant cells ( Fig . 3B–C , quantified in 3F ) . Similarly , knockdown of Raptor , a component of TORC1 complex , also significantly suppressed apoptosis in axnS rbf double mutant clones ( Fig . 3D–E , quantified in G ) . These results suggest that inactivation of axn leads to increased TORC1 signaling activity , which contributes to synergistic cell death in conjunction with rbf mutation . Deregulated activation of TORC1 by Tsc1 or Tsc2 mutation causes an imbalance between the metabolic demand and supply , and the Tsc1/Tsc2 mutant cells are highly dependent on glutamine metabolism for survival during energy stress [29] . Similarly , rbf mutants were found to exhibit altered glutathione metabolism and are hypersensitive to energy stress induced by fasting [15] . The observed effect of axn mutant on TORC1 signaling prompted us to test whether axn127 mutant larva also show hypersensitivity to fasting . Interestingly , axn127 mutant larva are much more sensitive to fasting than the controls ( FRT 82B ) and addition of glutamine to PBS largely suppressed the observed sensitivity of the axn127 mutants ( Fig . 4A ) . The increased sensitivity of axn127 mutant to fasting correlated with increased sensitivity to fasting-induced cell death in axn127 mutant clones , which is enhanced by lkb1 mutation ( Fig . 4C–G ) . LKB1 is a kinase that functions to balance cellular energetic needs and supply through AMPK-dependent and-independent pathways . Inactivation of LKB1 has been shown to increase death of cells under energy stress [9] . Taken together , these observations suggest that axn127 mutants have an altered metabolic process and show increased sensitivity to energy stress induced by fasting . Consistent with this notion , genome-wide expression studies using third instar axn127 homozygous mutants showed that the top functions affected in axn127 mutant include genes involved in metabolism , oxidation-reduction , stress response , signal transduction , and developmental processes ( Supplemental Tables S1 , S2 ) . We generate axn127 rbf120 double mutants to further test whether rbf axn mutations show synergistic effects in hypersensitivity to fasting . rbf120 is a viable weak rbf allele [30] . Consistent with previous reports [15] , more rbf120 larvae died than WT control after 28 hour fasting . axn127 larvae were also more sensitive than WT control to fasting . Interestingly , rbf120 axn127 double larvae were even more sensitive to fasting than either axn127 or rbf120 single mutants ( Fig . 4B ) . To test if excessive metabolic and energy stress contribute to the synergistic cell death of axn rbf double mutants similar to that observed for the rbf tsc2 mutant cells , we first determined whether axn single and axn rbf double mutant cells were under energy stress . The ATP/ADP ratio of eye discs with axn , axn rbf , or axn rbf Rheb mutant clones in Minute background were determined . Compared to wild-type cells ( FRT 82B ) , axn mutant cells had slightly lower ATP/ADP ratio ( P<0 . 01 ) , suggesting that they were under mild energy stress . The ATP/ADP ratio of axn rbf mutant cells was significantly lower than that of the axn mutant cells ( P<0 . 0001 ) , indicating that the double mutant cells were under severe energy stress ( Fig . 4L ) . Interestingly , blocking TORC1 activation by Rheb mutation increased the ATP/ADP ratio of axn rbf mutant cells to a level similar to that of the axn mutants ( Fig . 4L , Fig . S4 , p = 0 . 4 , between axn and axn rbf rheb ) , suggesting that inhibition of TORC1 activity decreased energy stress of the axn , rbf mutants . We further tested whether lkb1 mutation showed synergistic effects with axn or axn rbf . Although lkb1 single mutant did not show significant levels of apoptosis , lkb1 mutation induced synergistic cell death with axnS mutation and lkb1 axn rbf triple mutant cells had very high levels of cell death ( Fig . 4G–J , quantified in 4M ) . Taken together , these results suggest that axn mutants are under energy stress and require the LKB1 pathway for survival . In addition , it is likely that excessive metabolic stress of axn , rbf mutants contributes to the synergistic cell death . Hid is a critical regulator of apoptosis in Drosophila imaginal discs , and is induced by diverse developmental and stress signals including cell competition and DNA damage [31] , [32] . Rbf-E2f1 directly regulates Hid expression [8] , [33] . However the upregulated Hid expression and Hid protein level in rbf mutant clones were relatively weak and limited to the stripe just anterior to MF where rbf apoptosis occurs ( Fig . 5A , 5D ) . Mutation of axn127 alone did not affect Hid transcription or Hid protein levels ( Fig . 5B , 5E ) . Interestingly , significantly expanded Hid transcription and Hid protein were observed in axn127 rbf double mutant clones anterior to the MF ( Fig . 5C , 5F ) , which correlated with the observed synergistic apoptosis of these cells ( Fig . 1G and N ) . We further tested the effect of strong axn alleles on Hid . Both Hid expression and Hid protein were significantly upregulated in axnS as well as in axnS rbf double mutant clones , however it is difficult to tell if the Hid expression is synergistically upregulated in the double mutant clones ( Fig . 5J , L and data not shown ) . To determine if Hid induction contributes to the synergistic cell death observed in axn , rbf double mutant clones , axn , rbf mutant clones were induced in the hid mutant background . As shown in Fig . 5G–I , mutation of hid largely blocked apoptosis of the axnS , rbf double mutant cells , demonstrating the critical role of Hid induction to synergistic cell death of axnS , rbf mutant cells . Since blocking TORC1 activity blocks synergistic apoptosis of axn , rbf mutants , we tested the effect of inhibiting TORC1 on Hid induction . We observed that inhibiting TORC1 signaling by a rheb mutation strongly blocked induction of Hid transcription as well as accumulation of Hid protein ( Fig . 5J–M′ , white arrowheads ) , suggesting that induction of Hid in axn , rbf mutant clones is TORC1 dependent . Since TORC1 activity significantly alters cellular metabolic and energetic demand and supply and inhibition of TORC1 helps to restore the energy balance in axn , rbf mutant cells ( Fig . 4L ) , these results suggest that Hid induction and apoptosis in axn , rbf mutant cells is regulated , at least in part , by metabolic and energy stress , similar to the synergistic cell death of tsc2 , rbf mutant cells . The Rb/E2F and the Wnt signaling pathways are highly conserved between fly and mammalian systems . To determine whether deregulated Wnt signaling and Rb inactivation can also induce synergistic cell death in mammalian cells , we first determined whether activation of Wnt signaling can induce cell death in DU145 cells , a Rb mutant prostate cancer cell line [34] . Knockdown of Wnt signaling negative regulator APC using shRNA constructs strongly reduced the level of APC protein as shown by antibody staining ( Fig . 6A ) and increased the Wnt signaling reporter activities ( Fig . 6B , Supplementary Fig . S5A ) . To determine whether deregulation of Wnt signaling by APC knockdown induced cell death , we stained cells with an early apoptosis marker Annexin V together with the nucleic acid dye propidium iodide . We observed that depletion of APC significantly increased cell death in DU145 cells ( Fig . 6C , Supplementary Fig . S5B ) . In addition , knockdown of APC in DU145 cells significantly decreased viable cell numbers ( Fig . 6D , Supplementary Fig . S5C ) , and decreased the colony growth in soft agar ( Fig . 6E , Supplementary Fig . S5D ) . To determine whether the observed shAPC-induced death depends on the absence of Rb function , we investigated the effect of expressing WT Rb in APC knockdown DU145 cells . Expression of the transduced WT Rb can be easily detected ( Fig . 6F ) . Expression of WT Rb significantly decreased APC knockdown-induced death ( Fig . 6G ) , and partially restored the total viable cell numbers ( Fig . 6H ) . Taken together , these results demonstrate that knockdown of APC significantly induced the cell death , which is dependent on the absence of Rb function . Colorectal cancer cells commonly have deregulated Wnt signaling and intact Rb/E2F pathway [35] . Consistent with a previous report [36] , knockdown of Rb in HCT116 colorectal cancer cells leads to decreased Wnt signaling reporter activity ( Fig . 6I–J , Supplementary Fig . S5E–F ) and increased cell death ( Fig . 6K , Supplementary Fig . S5G ) . Rb knockdown-induced cell death in colorectal cancer cells was attributed to the reduced Wnt signaling activity [36] . To determine whether Rb knockdown induced cell death in HCT116 cells was due to reduced Wnt signaling or due to synergistic cell death induced by deregulated Wnt signaling and Rb inactivation , we set to distinguish these two possibilities in cells with depleted APC . Knockdown of APC significantly increased the Wnt signaling in HCT116 cells ( Fig . 6J , Supplementary Fig . S5F ) , indicating that APC significantly inhibited Wnt signaling even though these cells contain a β-catenin gain of function mutant allele . Importantly , Wnt signaling reporter activity was higher in APC and Rb double knockdown cells than that in control knockdown cells ( Fig . 6J , Supplementary Fig . S5F ) . However , increased Wnt signaling in the double knockdown cells did not suppress Rb knockdown-induced cell death . In fact , the cell death in Rb and APC double knockdown cells was even higher than those of the single or control knockdown cells ( Fig . 6k , Supplementary Fig . S5G ) . Therefore , although Rb depletion decreases Wnt signaling activity in colorectal cancer cells , its induction of cell death is likely mediated by the synergistic death effect from pRb inactivation and deregulated Wnt signaling . Synergistic cell death from inactivated Rb and deregulated Wg signaling in Drosophila depends on upregulated TORC1 activity ( Fig . 3 ) . To determine whether TORC1 activity also contributes to the synergistic cell death in mammalian cells , we determined the effect of inhibiting mTORC1 activity using rapamycin . Rapamycin potently blocked APC knockdown induced cell death in Rb mutant DU145 cells as well as Rb knockdown induced cell death in HCT116 cells ( Fig . 7A , B ) . These observations suggest that , similar to Drosophila , TORC1 activity is required for synergistic cell death induced by Rb inactivation in conjunction with deregulated Wnt signaling in mammalian cells . Our previous studies have shown that inactivation of Rb and TSC2 , a negative regulator of TORC1 , induced synergistic cell death in cancer cells through induction of excessive cellular stress , including oxidative stress [10] . We used DHE , a dye that detects superoxide , to determine whether oxidative stress is also associated with deregulated Wnt signaling and Rb inactivation induced cell death . As shown in Fig . 7 , highly elevated levels of DHE fluorescence were observed in APC-knockdown DU145 cells as well as in Rb-knockdown HCT116 cells grown in soft agar ( Fig . 7C–D ) . Furthermore , Rapamycin , which inhibits TORC1 activity , suppressed the ROS level in these knockdown cells ( Fig . 7E–F ) . Finally , NAC , a ROS scavenger , strongly rescued the knockdown-induced colony growth defects in soft agar ( Fig . 7G–H ) . Taken together , these observations suggest that Rb inactivation and deregulated Wnt signaling induced cell death requires TORC1 activity and involves oxidative stress induction .
This study revealed a novel and evolutionarily conserved synthetic lethal interaction between hyperactivated Wnt signaling and inactivated Rb . We show that a weak allele of axn with deregulated Wg signaling significantly alters the expression of metabolic genes and is hypersensitive to metabolic stress induced by fasting in Drosophila . Furthermore , we observe that hyperactivation of Wg signaling significantly increased TORC1 activity and induced excessive energy stress and synergistic cell death in conjunction with rbf mutation . These observations are consistent with the previous studies , which showed increased TORC1 activity by tsc1 or tsc2 mutation induced synergistic apoptosis with Rb mutation [10] , [11] . Our previous studies showed that mutation of lkb1 , a key regulator of energy metabolism under energy stress conditions , promoted synergistic cell death with rbf tsc1 mutations [9] . Similarly , we show here that axn rbf cells are also energy deficient and lkb1 mutation strongly enhanced the apoptotic effects of axn or rbf axn mutants . Interestingly , inhibition of TORC1 activity significantly suppressed synergistic cell death induced by deregulated Wg signaling and rbf inactivation , which correlated with decreased energy stress and decreased induction of apoptotic regulator Hid . These results provide further evidence that excessive metabolic and energetic stress contributes to the synergistic cell death . Finally we demonstrate that the phenotypes and mechanisms of axn rbf synergistic apoptosis in Drosophila are conserved in mammalian cells and that inactivation of Rb and APC induces synergistic cell death that requires TORC1 activity and involves oxidative stress induction . Wnt/Wg signaling is one of the key developmental signaling pathways repeatedly used in different developmental settings to regulate cell proliferation , apoptosis , as well as cell differentiation . The consequence of deregulated Wnt signaling depends on particular cellular contexts . In Drosophila larval eye discs , Wg signaling is essential for proliferation of the progenitor cells anterior to the MF . Mutant clones of axn127 , which does not affect cell type specification or patterning , showed Hid upregulation and synergistic cell death with rbf only in the anterior proliferating region . In contrast , strong axn alleles , which blocks photoreceptor differentiation , caused synergistic cell death with rbf in both the anterior and the posterior clones . Therefore , it appears that synergistic cell death of deregulated Wnt signaling and rbf inactivation is mainly observed in the proliferating progenitor cells . Consistent with this , we found that the observed synergistic cell death is associated with increased TORC1 activity , metabolic stress , and cell proliferation . In mammalian systems , Wnt signaling plays important roles in maintaining stem cell and progenitor cell homeostasis and deregulated Wnt signaling is observed in many types of cancers , particularly the colorectal cancers . It is quite likely that synergistic cell death interactions between deregulated Wnt signaling and inactivated Rb potentially play important roles in maintaining stem cell homeostasis as well as during cancer development . While Wnt signaling is required to maintain intestine stem cells , hyperactivation of Wnt signaling results in increased cell proliferation as well as increased apoptosis [37] , [38] . Similarly inactivation of APC in hematopoietic stem cells ( HSCs ) increases cell proliferation as well as apoptosis , leading to HSC exhaustion and bone marrow failure [39] . Since pRb is inactivated during G1/S transition , pRb is partially inactivated as these stem cells or progenitor cells proliferate . An interesting possibility is that different levels of Wnt signaling activation or pRb inactivation will cause graded levels of metabolic alterations . When combined Wnt signaling hyperactivation and pRb inactivation induced metabolic change past a certain threshold , excessive metabolic stress and cell death will be induced . It is interesting to note that although Rb inactivation is found in almost half of cancer cells , colorectal cancers often show Rb copy gains with high level of Rb expression [35] . Since deregulated Wnt activities is the key cancer initiating event that exists in almost all colorectal cancer cells , the high Rb level can potentially prevent cell death induced by hyperactivated Wnt signaling , particularly during early cancer progression . In addition to inducing synergistic cell death with deregulated Wnt signaling , high E2F activities were also found to antagonize Wnt signaling by degrading β-catenin in a GSK3β independent manner [36] . It is possible that the Rb-E2F and Wnt signaling pathway may crosstalk at multiple levels , and Wnt signaling can induce either pro-apoptotic or survival signals depending on particular cellular context . The observed synergistic cell death between hyperactive Wnt signaling and inactivated Rb may also contribute to the cancer cells drug sensitivity . A recent study showed that upregulation of Wnt signaling is required for cell death induction in melanoma cells by PLX4720 , a selective inhibitor of activated BRAF ( V600E ) . PLX4720 increased Wnt signaling and induced Bim expression and cell death in A375 melanoma cells , which was blocked by β-catenin ( CTNNB1 ) siRNA [40] . A375 cells have lost the expression of p16INK4a , which is a cyclin dependent kinase ( CDK ) inhibitor that regulates the phosphorylation of pRb by D-type CDKs [41] . Therefore , pRb is likely at least partially inactivated in these cells . Interestingly , analysis of the Genomics of Drug Sensitivity in Cancer database [42] , a publicly available IC50 dataset of 147 anticancer agents on over 1000 tumor cell lines , revealed that PLX4720 was one of the seven drugs that show increased effectiveness toward cancers that have genomic alterations of the Rb gene [43] . Therefore , it will be interesting to investigate whether Wnt induced apoptosis in A375 cells requires Rb inactivation . Deregulated TORC1 activity is often observed in cancers and inhibition of TORC1 activity can potentially be used as a strategy to inhibit cancer growth . However , the clinical trials of the TORC1 inhibitor Rapamycin and its derivatives have only seen very limited success in small subset of cancers [44] . Besides the possibilities that these inhibitors are not potent enough to completely inhibit TORC1 or they activate feedback signaling , our studies raise the possibility that inhibition of TORC1 decreases the stress levels in cancer cells and promotes cancer cell survival . Indeed , decreasing the activities of TORC1 or its downstream target S6K partially rescues the Rb- TSC synergistic cell death [10] , [11] . Several studies described how increased Wnt signaling activates TORC1 activity . One possible mechanism is mediated by the inhibition of mTOR by GSK3 through the phosphorylation of TSC2 [45] , [46] , [47] , [48] , [49] . In this case , increased Wnt signaling will activate mTOR by inhibiting GSK3 . Another mechanism described recently is that GSK3 and mTOR cooperate to regulate S6K phosphorylation [50] . Additionally , canonical Wg signaling has been shown to promote insulin sensitivity by upregulating insulin receptor expression [51] . Therefore , Wnt and TOR signaling pathways intersect at multiple levels .
Fly stocks used in this study include: rbf15aΔ [8] , dtsc129 [52] , lkb1X5 [53] , hid138 [8] . axnEY10228 ( BL17649 ) , axnE77 [21] , axnS044230 [54] , APC1Q8 APC279 [54] , Hid-lacz [55] , Rheb4L1 ( BL39737 ) , UAS-Axn-GFP ( BL7224 ) , UAS-Raptor RNAi ( BL34814 ) , UAS-Rbf RNAi ( BL36744 ) , UAS-ArmS10 ( BL 4782 ) , aos-lacz ( BL2513 ) , UAS-TCFDN ( BL4785 ) , UAS-RasV12 ( BL4847 ) , UAS-GSK3DN , PCNA-GFP [56] . Ethyl methanesulfonate ( EMS ) -induced screen to identify mutations that can modulate the phenotypes of rbf was carried similar as described [8] , except that w; p{ry+ , neoFRT82B} males were used for mutagenesis , and rbf15aΔ , w , eyFLP; p{ry+ , neoFRT82B} p{w+ , Ubi-GFP} p{w+ , rbf-G3} and w , eyFLP; p{ry+ , neoFRT82B} p{w+ , Ubi-GFP} stocks were used for screening and rbf dependence test . Total RNA was isolated with TRIzol ( Invitrogen ) . cDNA was synthesized with 1 µg total RNA , M-MLV Reverse Transcriptase ( Invitrogen ) , and 3′RACE-T7 primer ( 5′-TAATACGACTCA CTATAGGGTTTTTTTTTTTTTTTTTTTTTTTV-3′ ( V = A , G , or C ) ) . Nested PCR was first performed with the Axin3′middleF primer ( 5′-CGGGTGTGGAAGGACCAAA-3′ ) and T7 primer ( 5′-TAATACGACTCACTATAGGG-3′ ) , and then the Axin3′middleF-2 primer ( 5′-ATTCCGGAATGGTCAGCGA-3′ ) and T7 primer . PCR products were gel purified and sequenced . Immunostaining was performed at room temperature unless indicated otherwise . Larval imaginal discs were dissected in 1× PBS , fixed with 4% formaldehyde in PBS for 25 min , washed twice with 1× PBS with 0 . 3% Triton-X100 ( PBST ) , and incubated with primary antibody in blocking solution ( PBST plus 5% normal goat serum ) overnight at 4°C . Primary antibodies used: rabbit anti-activated Caspase-3 ( C3 , 1∶300 from Cell Signaling ) , mouse anti-β-Galactosidase ( 1∶100 , DSHB ) , rat anti-ELAV ( 1∶50 , DSHB ) , Guinea pig anti-Senseless [57] , and Guinea pig anti-E2F1 ( Orr-Weaver lab ) . Guinea pig anti-Hid antibody was affinity purified with recombinant GST-Hid [58] . Following incubation with primary antibody , samples were washed three times ( 10 minutes each ) in PBST , and incubated with secondary antibodies from Jackson ImmunoResearch ( 1∶200 to 1∶400 ) . Sample was mounted in 70% Glycerol with 1 , 4-diazabicyclo[2 . 2 . 2]octane ( DABCO ) at 12 . 5 mg/mL . For mammalian cell staining , infected cells were seeded onto glass coverslips , and processed for staining . Fixed , permeabilized , and blocked cells were incubated with rabbit anti-APC M2 ( kindly provided by Kristi Neufeld , University of Kansas ) , followed by FITC-coupled secondary antibody . Imaging was done with the Zeiss Axioscope/ApoTome microscope using the AxioCam CCD camera controlled by Zeiss Axiovision software . In experiments with internal controls ( for example , the WT tissues from the same disc that do not show cell death ) , the exposure time for each sample were determined using the “measure” function in Axiovision for each channel to get optimal exposure without signal oversaturation . For experiments with no internal controls , exposure time was fixed using the genotype with brightest signal to avoid overexposure . Cell death ( % ) is determined as described previously [10] by the percentage of clone area ( pixels ) that have above background level of caspase 3 ( C3 ) signal using the Histogram function in Photoshop . The background level of C3 signal was determined as the level that is equal or below 99% of the C3 signal in the WT tissues that have no apoptosis . The Average and standard deviation of percent cell death for each genotype discs was then determined and compared . 40 Drosophila eye discs with each specific genotype were dissected in insect cell media CCM3 ( Hyclone ) , and moved to 1 . 5 ml tubes with 100 µl 1× SDS-PAGE loading buffer immediately . The samples were pipetted for several times , boiled for 5 minutes , quickly centrifuged , and 20 µl of them were loaded for SDS-PAGE . For mammalian samples , cells were washed twice with 1× PBS , and lysed in RIPA buffer ( 50 mm Tris . Cl [pH 7 . 4] , 150 mm NaCl , 2 mm EDTA , 1% NP40 , 0 . 1% SDS , 0 . 5% sodium deoxycholate , plus protease inhibitors ) . Primary antibodies are rabbit anti-pS6K ( 1∶300 , Cell Signaling ) , mouse anti β-actin ( 1∶1000 Santa Cruz ) , and mouse anti-Rb 4 . 1 ( 1∶10 , Developmental Studies Hybridoma Bank ) . The goat anti-mouse IgG and goat anti-rabbit IgG secondary antibodies were obtained from Li-Cor . Western detection was carried out using a Li-Cor Odyssey image reader . Eye imaginal discs with specific genotypes were dissected , pipetted with 120 uL 1× Passive Lysis Buffer ( Promega ) for 15 times in a 1 . 5 mL tube on ice , boiled for 5 minutes , then incubated on ice for 2 minutes . After centrifugation at 18 , 000G for 2 minutes , 20 uL of each sample was used to assay the ADP∶ATP ratio using the Enzylight kit according the manufacturer's protocol ( BioAssay Systems ) . To induce metabolic stress , FRT 82B Axn127 and FRT 82B control 2nd instar larvae were collected at 72 hour after egg lay , rinsed to remove any residual fly food , and transferred into empty vials containing one 11 cm by 21 cm Kimwipe soaked with 1 ml of either 1× PBS or PBS with glutamine . Eight vials containing 25 larvae each were used per genotype per condition . These vials were incubated at 25°C for 48 hours , at which point the Kimwipe was extracted and the larvae were characterized . Drosophila larvae were determined to be viable if they responded to stimuli from poking with a blunted pair of forceps . For eye disc starvation , dissected eye discs were left in 1× PBS at room temperature for 3 hours before fixation , and eye discs fixed immediately after dissection were used as control ( 0 hr ) . The immunostaing with C3 antibody is the same as described above . Larvae of Oregon R Drosophila ( control ) and axn127 homozygous mutants ( w1118; +; FRT82B axn127 ) were collected at the third instar wandering stage . Total RNA was extracted from three larvae per sample with 1 . 0 ml of TRIzol Reagent ( Life Technologies Corporation ) according to the manufacturer's instructions . The microarray analysis was performed according to the protocol that was described previously [59] . The complete sets of microarray data have been deposited in the ArrayExpress database ( http://www . ebi . ac . uk/arrayexpress/; accession number is E-MTAB-2342 ) . Gene Ontology was performed with GO-TermFinder ( http://amigo . geneontology . org/cgi-bin/amigo/term_enrichment ) [60] . DU145 and HCT116 cells were obtained from the American Type Culture Collection . All the cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum ( FBS , Atlas Biologicals ) , 50 IU penicillin/streptomycin , and 2 mmol/l L-glutamine ( Invitrogen ) in a humidified atmosphere with 5% CO2 at 37°C . Human pRb was subcloned into the lentiviral expression vector pCDHCMV-EF1-puro ( System Biosciences ) . The pLKO . 1 lentiviral RNAi expression system was used to construct lentiviral shRNA . The sequences of shRNA used in this study included the following: shAPC-1: 5′-CCGGTGAGGTCATCTCAGAACAAGCTCGAGCTTGTTCTGAGATGACCTCtttttt-3′ shAPC-2: 5′-CCGGTAAGACGTTGCGAGAAGTTGGACTCGAGTCCAACTTCTCGCAACGTCTTtttttt-3′ shpRb-1: 5′-CCGGCGACGAGTCAAACAAGCCAATCTCGAGATTGGCTTGTTTGACTCGTCGTTTTTG shRb-3: 5′-CCGGTGGTTGTGTCGAAATTGGATCACTCGAGTGATCCAATTTCGACACAACCTTTTTT-3′ shGFP: 5′-CCGGTACGTCTATATCATGGCCGACAACTAGTTGTCGGCCATGATATAGACGTTTTTTG-3′ The shGFP was used as a control in this study . Viral packaging was done according to the previously described protocol [10] . Briefly , expression plasmids pCMV-dR8 . 91 and pCMV-VSV-G were cotransfected into HEK293T cells using the calcium phosphate method at 10∶5∶5 µg ( for a 10-cm dish ) . The transfection medium containing calcium phosphate and plasmid mixture was replaced with fresh complete medium after incubation for 6 hr . Media containing virus was collected 48 hr after transfection , and then concentrated at 19 , 400 g for 2 hr . The virus pellet was re-dissolved , and stocked at −80°C . Cells were infected with the viruses for 48 hr , and were treated as described . Quantification of cell death was performed using FACSCanto ( BD Biosciences ) after cells were stained with Annexin V-FITC ( BD Biosciences ) and propidium iodide ( Sigma ) according to manufacturer's specifications . Rapamycin rescue assays were performed in the presence of 20 ng/ml Rapamycin or vehicle control . Cells were treated with lentivirus as described above , and were plated into a 24-well plate , followed by transfection by lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instruction . Each transfection contained 800 ng of TOPflash-luc or FOPflash-luc , and 5 ng of phRL-Luc . Cell extracts were prepared 48 hrs post-transfection , and the luciferase activity was measured using Dual Luciferase Reporter Assay System ( Promega ) according to the manufacturer's instruction . Luciferase activity was read on a BD Monolight 3010 Luminometer . All data points presented are the average measurement of three independent transfections . For growth assay , 104 cells suspended in 0 . 35% agarose solution were poured over hard-bottomed agar ( 0 . 6% ) previously solidified in 6-well plates . Cells were cultured in a humidified atmosphere with 5% CO2 at 37°C for 3–4 weeks , and then colonies were counted . Soft agar growth rescue assays were performed in the presence of 10 mM NAC or vehicle control added to the top layer mix at the time of plating . For ROS assay , 105 cells were seeded between top agar layer and bottom agar layer for 16 hrs , and then 1 ml of complete medium containing 20 µM of DHE was added onto the top agar layer . After incubation for 1 hr , the medium was aspirated and the top agar layer was carefully removed . Cells were processed for imaging with a Zeiss fluorescence microscope . | Inactivation of Rb tumor suppressor is common in cancers . Therefore , identification of genes and pathways that are synthetic lethal with Rb will provide new insights into the role of Rb in cancer development and promote the development of novel therapeutic approaches . Here we identified a novel synthetic lethal interaction between Rb inactivation and hyperactivated Wnt signaling and showed that this synthetic lethal interaction is conserved in mammalian systems . We demonstrate that hyperactivated Wnt signaling activate TORC1 activity and induce excessive energy stress with inactivated Rb tumor suppressor , which underpins the evolutionarily conserved synthetic lethal interaction . This study provides novel insights into the interactions between the Rb , Wnt , and mTOR pathways in regulating cellular energy balance , cell growth , and survival . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"oncology",
"developmental",
"biology",
"medicine",
"and",
"health",
"sciences",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"cell",
"biology"
] | 2014 | Hyperactivated Wnt Signaling Induces Synthetic Lethal Interaction with Rb Inactivation by Elevating TORC1 Activities |
Dengue is a fast spreading mosquito-borne disease that affects more than half of the population worldwide . An unprecedented outbreak happened in Guangzhou , China in 2014 , which contributed 52 percent of all dengue cases that occurred in mainland China between 1990 and 2015 . Our previous analysis , based on a deterministic model , concluded that the early timing of the first imported case that triggered local transmission and the excessive rainfall thereafter were the most important determinants of the large final epidemic size in 2014 . However , the deterministic model did not allow us to explore the driving force of the early local transmission . Here , we expand the model to include stochastic elements and calculate the successful invasion rate of cases that entered Guangzhou at different times under different climate and intervention scenarios . The conclusion is that the higher number of imported cases in May and June was responsible for the early outbreak instead of climate . Although the excessive rainfall in 2014 did increase the success rate , this effect was offset by the low initial water level caused by interventions in late 2013 . The success rate is strongly dependent on mosquito abundance during the recovery period of the imported case , since the first step of a successful invasion is infecting at least one local mosquito . The average final epidemic size of successful invasion decreases exponentially with introduction time , which means if an imported case in early summer initiates the infection process , the final number infected can be extremely large . Therefore , dengue outbreaks occurring in Thailand , Singapore , Malaysia and Vietnam in early summer merit greater attention , since the travel volumes between Guangzhou and these countries are large . As the climate changes , destroying mosquito breeding sites in Guangzhou can mitigate the detrimental effects of the probable increase in rainfall in spring and summer .
As the most important mosquito-borne disease globally , the incidence of dengue has increased 30-fold in the past 50 years [1] . In 2013 alone , dengue was responsible for 576 , 900 years of premature life lost and 1 . 14 million disability-adjusted life-years globally [2] . However , dengue was not a serious concern in mainland China before 2014 . There were only 73 , 179 cases from 1990 to 2015 , among which 47 , 056 ( 64 . 3% ) occurred in 2014 , and 80 . 8 percent of these cases were contributed by Guangzhou [3] . Because of its warm and wet climate , as well as the close ties with dengue endemic Southeast Asian countries , Guangzhou has a high risk of local dengue transmission . The introduction of dengue virus ( DENV ) by imported cases is required annually for local epidemics in Guangzhou , since the dengue virus cannot disseminate to the salivary glands of Aedes albopictus when temperature is below 18°C [4] and adult mosquitoes rarely survive the winter . Many studies have attempted to explain the unprecedented dengue outbreak in Guangzhou , 2014 . Possible explanations include more imported cases , higher mosquito abundance due to a more favorable climate , higher larval development and adult survival rates caused by urbanization , and lack of diagnostic experience [5 , 6] . The result of our earlier analysis using a deterministic mathematical model identified the early timing of local transmission and climate as the most important determinants of the large final epidemic size ( FES ) [7] , but the model could not explain why local transmission occurred earlier that year . Because a stochastic model can produce different results every run under the same conditions , an imported case can lead to local transmission in some runs but not so in others . Therefore , a stochastic model can shed more light on the outbreak probability and can provide greater insight into the various explanations of earlier local transmission , such as a more favorable climate for mosquito growth and more imported cases . Clearly , understanding the reasons underlying the early and large outbreak can help to prevent dengue outbreaks in the future , and reduce potential economic and health impacts . Stochastic models have been widely used to study the invasion of a disease into non-endemic areas , such as dengue virus transmitted by Ae . aegypti mosquitoes [8] , dengue virus in Madeira [9] , and Chikungunya virus in the United States [10] . However , none of these models considered the importance of water availability in restricting the environmental carrying capacity and density-dependent rates . In this paper , we simplify our deterministic model by leaving out vertical transmission from infected adults to eggs since it was found to be unimportant in the earlier analysis [7] , and then extend it to incorporate stochastic effects . A hybrid deterministic/stochastic algorithm originally used in simulating chemical reactions [11–13] is used to simulate the dengue transmission dynamics . This algorithm can take advantage of the low computational burden of deterministic models while still incorporating stochastic effects . Six different scenarios are constructed to explore the determinants of early outbreak in 2014 and the effectiveness of interventions ( insecticide spraying and pooled water removal ) in reducing dengue transmission risk . We also address the determinants of the outbreak size and FES , as well as the implications for prediction and prevention .
The Ethics Committee of the Guangzhou Center for Disease Control and Prevention reviewed and approved this study . All patient data was made anonymous prior to the analysis . Guangzhou , with a population of 13 . 1 million at the end of 2014 [14] , is the largest city in South China . The climate is characterized by warm and humid summers and mild and dry winters . The climate data downloaded from China Meteorological Data Sharing Service System ( CMDSSS ) indicates that the annual mean temperature from 1985 to 2014 was 22 . 4°C . January had the lowest average temperature of 13 . 7°C , while July and August had the highest at 28 . 9 and 28 . 7°C , respectively ( Fig 1A ) . The average annual accumulated precipitation was 1 , 821 mm , of which 1 , 490 mm ( 81 . 9% ) occurred between April and September ( Fig 1B ) . The warm and wet summer is favorable for the growth of Ae . albopictus , the sole vector of dengue transmission in Guangzhou [15] . Because low temperature is unable to support virus replication in Ae . albopictus in the winter [4] , the occurrence of dengue epidemics in Guangzhou depends highly on imported cases from surrounding endemic countries , such as Thailand , Singapore , Malaysia , Philippines , Vietnam , Cambodia , and Indonesia . Serving as a transportation hub for these countries increases the outbreak risk of Guangzhou further ( Fig 1C ) . We collected daily reported dengue case numbers in 2013 and 2014 from the Guangzhou Center for Disease Control and Prevention ( Guangzhou CDC ) . These data are also available from the Health and Family Planning Commission of Guangdong Province ( http://www . gdwst . gov . cn/ ) in the transmission season . Both passive and active surveillance systems are used in China . Health-care providers are required to diagnose the disease according to the National Diagnostic Criteria for Dengue Fever ( WS216-2008 ) and then report the new cases to the web-based National Notifiable Infectious Disease Reporting Information System within 24 hours . After the confirmation of a case , active detection is initiated and conducted by various means , such as interviewing residents in the same community , checking employee attendance records in the same work place , and checking the outpatient medical records in nearby health facilities [16] . Cases are categorized further into imported and indigenous cases based on the travel and mosquito biting history . Imported cases are those who have traveled to dengue endemic regions and been bitten by mosquitoes less than 15 days before the symptom onset [6] . This dataset was used to calibrate the deterministic model . Entomological surveillance data , including the Breteau index ( BI ) and mosquito ovitrap index ( MOI ) for 2013 and 2014 , were also obtained from Guangzhou CDC . BI is a representation of larva abundance , which is defined as the number of water containers infested with larva per 100 houses inspected , while MOI is a proxy for adult abundance defined as the percentage of positive ovitraps in a specific area . The daily temperature , precipitation and evaporation data for Guangzhou from 1985 to 2014 were downloaded from CMDSSS ( http://data . cma . cn/ ) . The 30-yr daily average climatic factors were calculated to represent a typical year and were used in Scenario 2014 to Avg described later . Climate datasets were used as inputs to the model . Human population data were extracted from the Guangdong Statistical Yearbook [17–19] to estimate the birth and death rates , which were treated as known parameters in the deterministic model . In addition , the initial value for the susceptible human population in the model was set to the population at the end of 2011 , because we assumed that all residents were susceptible to dengue virus infection since no big outbreaks had occurred in Guangzhou from 1978 . To investigate the tourist exchange between Guangzhou and dengue endemic countries , we also extracted the number of direct air travelers between Guangzhou and different countries from on-flight origin and destination ( OFOD ) data provided by International Civil Aviation Organization ( ICAO ) ( https://www4 . icao . int/NewDataPlus/ ) . Since this dataset leaves out indirect air travelers or travelers by sea or by land , we also collected the number of foreign tourists staying overnight in Guangzhou and the number of tourists who traveled from Guangzhou with tour groups from the Tourism Administration of Guangzhou ( http://www . gzly . gov . cn/index . html ) as a complement ( S1 Table ) . To simulate the population dynamics and dengue virus infection status of both Ae . albopictus and humans , we constructed a compartment model shown in Fig 2 which is based on Ross-Macdonald model [20 , 21] . Several modifications were made to the basic model to emphasis the influence of temperature and precipitation , such as modelling water level and aquatic stages explicitly , and using temperature- and density-dependent rates . The model has been described in detail elsewhere [7] . Because vertical transmission was found to be unimportant in the 2014 outbreak [7] , here we removed the vertical infected egg , larva , pupa and emerging adult compartments to simplify the model structure and to reduce computational complexity . Immature life stages eggs ( E ) , larvae ( L ) , and pupae ( P ) were included to incorporate the effects of water availability and temperature on vector abundance . Emerging adults ( Aeu ) were also incorporated as a separate compartment because they do not bite . Then , according to infection status , the biting mosquito population was divided further into susceptible ( As ) , exposed ( Ae ) and infectious ( Ai ) adults , and the human population was divided into susceptible ( Hs ) , exposed ( He ) , infectious ( Hi ) and recovered ( Hr ) individuals . Transition rates in the model depend on temperature and water availability . Temperature can influence the development rate from eggs to larvae , larvae to pupae , and pupae to emerging adults ( Events 2 , 4 and 6 in Fig 2 ) , the mortality rate of larvae , pupae and adults ( Events 3 , 5 , 7 , 9 , 13 and 16 ) , the biting rate of adults ( Events 12 and 19 ) , the extrinsic incubation period ( EIP ) of dengue virus ( Event 14 ) , and the average number of eggs and duration of each gonotrophic cycle ( Events 10 , 11 and 15 ) [23–25] . Therefore , temperature-dependent functions whose form is based on [23 , 26] were used to describe these rates , and the coefficients in these functions were estimated from experiments conducted on Ae . albopictus strains from South China [25 , 27] . Furthermore , the development rate from eggs to larvae , larvae to pupae , and the mortality rate of larvae also depend on the larval density ( Events 2–4 ) , which further depend on the current available water level determined by rainfall , evaporation , the minimum and maximum water level ( ωmin and ωmax ) of the system , and interventions ( eg . emptying water containers ) [28] . Minimum water level represents standing waters that are difficult to evaporate , such as the water in containers with lids or in shaded areas , while the maximum water level is the upper limit of the water in the system beyond which water will overflow . Since these two parameters cannot be obtained from the literature or surveys , they were estimated by the deterministic model . More information for the model structure can be found in the S1 File . As in the previous paper [7] , we still include the arrival of imported cases , intervention , the spillover effect and diapausing eggs in the current model . The model was run from 2012 to 2014 , though we only fit the model to the observed daily reported cases of 2013 and 2014 . Year 2012 was included to get a reliable mosquito population for 2013 and 2014 , because the initial value for eggs can only affect the mosquito abundance of the first simulated year , but not in later years [29] . There are 19 parameters in our model and their values are highly uncertain . In addition , we are more confident about the pattern of the daily reported new cases and mosquito surveillance data than the exact value of these data on any given day . Therefore , instead of employing Markov Chain Monte Carlo ( MCMC ) fitting procedures common in epidemiologic mathematical modelling [9 , 30] , we used a parameter estimation strategy originally called regional sensitivity analysis ( RSA ) [31] to fit the model to the pattern of the surveillance data . The detailed process was described elsewhere [7] , and only a brief description is provided here . To calibrate the deterministic model , a broad range for each parameter was specified based on the literature or our best knowledge . Classification criteria were defined to describe whether the model output mimicked the observed timing and approximate number of reported cases at the onset , peak and end phase of the epidemics ( see details for the criteria in S1 File ) . A Monte Carlo procedure was then followed in which random samples were drawn from the specified parameter space and used to run the mode . The output of each realization was classified into one of two subgroups “pass” or “fail” according to whether the result met all of the classification criteria . After accumulating a sufficient number of parameter vectors in the “pass” group , the characteristics of the cumulative univariate distributions were used to trim the range for each parameter to get a smaller parameter subspace with a higher passing rate . A total of 5 trimming cycles were conducted until the pass and fail univariate distributions offered no further guidance on identifying subspaces with higher pass rates . The results are presented in S1 File . Unlike MCMC fitting , which gives only one parameter set with a confidence interval , RSA gives multiple parameter sets matching the passing criteria . Here we obtained 5 , 320 parameter sets out of 100 , 000 runs after the last trimming cycle ( Cycle 5 ) in contrast with the 83 passing sets from 800 , 000 trials initially ( Cycle 1 ) . Fig 3A shows the trajectories and the envelope of the number of daily new cases produced by the 5 , 320 sets in Cycle 5 . As expected , these passing parameter sets produce trajectories which mimic the pattern of the field data successfully , except that the peak number of cases in 2014 is a little lower and occurs earlier than observed . Although the observed initial exponential growth rate of 2014 is higher than that of 2013 , the model produces very similar values , presumably because we assume that the intrinsic incubation period , recovery period , transmission probability from vector to human and from human to vector , and reporting rate remain the same from year to year . This lack of fit suggests difference in dengue virus virulence or reporting rate between these 2 years , which may need further investigation in the future . The number of larvae and adult mosquitoes output by the model shows the same patterns as the mosquito surveillance data for both BI and MOI . Since these entomological datasets were not used in model calibration , they support the validity of our model and its parameterization . In transitioning from a purely deterministic model to one suitable for exploring stochastic effects , we must take account of the earlier finding that there are extended regions of the parameter space that led to good fits to the calibration data for the deterministic model . Since thousands of simulations were needed for each parameter set , it is impractical to use all the 5 , 320 passing sets of Cycle 5 . As a result , 100 parameter sets were sampled randomly to represent the space of good fitting vectors ( S2 Table ) . To reduce the computational costs further , a simulation procedure developed specifically for hybrid deterministic/stochastic models was utilized . The hybrid deterministic/stochastic algorithm can add stochasticity to the dengue transmission model and obtain estimates of the success invasion rate of one imported case and the distribution of FES . This method has been widely used in simulating chemical reactions . It can improve the calculation efficiency significantly while still producing a similar pattern as that achieved by using the exact method [11–13 , 32 , 33] . The transitions between different compartments here , analogous to the chemical reactions in those studies , were partitioned into “fast” and “slow” subsets according to their transition rates . Fast events happen frequently and have lower level of stochasticity , thus they are simulated by the deterministic model using ordinary differential equations ( ODEs ) ( Events 1–11 , 15 , 17 and 18 in Fig 2 ) , while slow events happen infrequently so they must be modeled stochastically ( Events 12–14 , 16 , and 19–24 ) . Since there were only small numbers of exposed and infected mosquitoes ( Ae and Ai ) , and of exposed , infected or recovered humans ( He , Hi and Hr ) at the beginning of the outbreak , the mortality rates of these compartments ( Events 13 , 16 , 20 , 22 , and 24 ) and the transition rates from these compartments ( Events 14 , 21 , and 23 ) were relatively low . Hence these events are considered as slow events . In addition , the transition rate of Event 12 ( bαhvHiAs/N ) , which is the mosquito infection via human contagion , is also low , because b is the temperature-dependent biting rate which ranges from 0 to 1; αhv is the transmission probability of dengue virus from infected human to susceptible adult mosquito which also ranges from 0 to 1; Hi is the number of infected humans which is a small number at the beginning of the outbreak; and As/N is the ratio of mosquitoes to humans , which is also small when compared with mosquito abundance or the human population . Therefore , Events 12 and 19 are slow events , for the same reason . For the stochastic simulation of the slow events , instead of the widely utilized Gillespie’s stochastic simulation algorithm ( SSA ) in the chemical studies , here we used the adaptive tau-leap algorithm [34–36] , which is more commonly used in epidemiological models to reduce the computational burden [10 , 34 , 35] . To test the validity of the tau-leap algorithm , Gillespie’s SSA was also tried for two randomly chosen parameter sets . The Kolmogorov–Smirnov two-sample test for both parameter sets indicated no significant difference between the FES distributions produced by different algorithms . Therefore , only the tau-leaping algorithm was applied in the following analysis . The time step tau was initially set to be 1/5 day , which is appropriate for a population of the order of millions [37] . If high numbers of events occurred in this time interval which led to negative population sizes , a new value of tau was adopted as tau/2 to shorten the time interval and avoid negative population sizes . The details of the implementation of the model are shown in S1 File . Dengue transmission risk is mainly determined by the mosquito-to-human ratio , temperature , and the immune status of human population [38] . Since the human population and its immune status do not vary much from year to year , the only differences between 2014 and other years were mosquito abundance and temperature . Mosquito abundance further depends on the availability of breeding sites , represented by water level in the model , and human interventions . The most common interventions in Guangzhou are insecticide spraying and emptying water containers , which can reduce the abundance of adults or aquatic stage of mosquitoes immediately . Emptying water containers can also affect the abundance through water availability and the density-dependent rate . Six different scenarios were designed here to investigate the role of climate and human interventions in determining the potential and FES of the dengue outbreak in 2014 ( Table 1 ) . The first scenario used the observed climate data of 2014 and serves as the baseline for comparison ( Scenario 2014 ) . Then , to study the role of climate only , we replaced the climate files of 2014 in Scenario 2014 with those of 2013 ( Scenario 2014 to 2013 ) and the 30-yr average ( Scenario 2014 to Avg ) . The 30-yr average was used here to represent the climate in a typical year . In addition to climate , year 2014 also differed from year 2013 in the initial water level , which was determined by the climate and interventions in the previous year . The initial water level in 2014 is lower than 2013 , because more water was removed in 2013 than in 2012 by more frequent interventions . The number of reported dengue cases in Guangzhou increased from 139 in 2012 , to 1 , 249 in 2013 , and then to 37 , 341 in 2014 , which led to increasing intervention frequency over these years . Therefore , we compared the results of Scenario 2013 and Scenario 2014 to determine the combined effect of climate and intervention . Moreover , aiming to understand the outbreak potential under only natural conditions , we also designed two additional scenarios without interventions ( Scenario 2014 w/o intervention and Scenario 2013 w/o intervention ) . The previous deterministic model [7] suggested that the interventions can reduce the FES effectively , so here these two scenarios were included to estimate the quantitative effects of intervention in reducing dengue transmission risk . Turning to the role of the timing of imported cases , the introduction date was varied between March 21st and November 26th ( the mosquito growth season ) at 10-day intervals ( a total of 25 different introduction dates ) . As with the deterministic model , the stochastic model was also run from Jan 1st , 2012 to December 31st , 2014 . Five hundred iterations were run for each combination of scenario and introduction date . The FES of the last simulated year was recorded for each iteration . Therefore , a total of 7 . 5 million simulations were conducted ( 100 sample sets , 25 introduction dates , 6 scenarios , and 500 repetitions ) . The successful invasion rate of an imported case was defined as the proportion of repetitions with a FES greater than 0 . The average final outbreak size only accounts for successful invasions , and can be calculated as the mean of FESs greater than 0 in 500 repetitions , while the average final epidemic size was the mean FES of all invasions . R 3 . 2 . 3 [39] was used for simulation , data analysis and visualization . The package deSolve was used for solving the ODEs [40] , Rcpp , foreach , and doSNOW for parallel computing to increase the computational speed [41 , 42] , and ggplot2 for data visualization [43] .
In order to determine whether the early outbreak in 2014 was caused by a particularly favorable climate for mosquito growth , we held all the other conditions constant , such as initial water level and the timing of interventions , but only changed the temperature , precipitation and evaporation of 2014 to those of 2013 and then to the 30-yr average ( Scenario 2014 to 2013 and 2014 to Avg ) . Fig 4A shows that if an imported case is introduced to the system on the same day in early summer ( the shaded rectangle ) , Scenario 2014 has the highest success rate . This implies that the climate in 2014 was more favorable for an early outbreak . For all the three scenarios , the success rate increases in late spring and early summer , remains at a high level from May to July , and then decreases until it reaches and stays at a low level in October . The biggest difference in the success rate between these scenarios occurs between May and July , the same period as the peak of success rate . Fig 4B shows the comparison between two scenarios with observed climate and intervention ( Scenario 2014 and 2013 ) and two scenarios with observed climate , but without intervention ( Scenario 2014 w/o intervention and 2013 w/o intervention ) . The result for Scenario 2014 is depicted here again as a baseline for comparison . Although the success rate of Scenario 2014 is higher than that of Scenario 2014 to 2013 and 2014 to Avg when only the climate is considered ( Fig 4A ) , it is lower than that of Scenario 2013 when the combined effect of lower initial water level and favorable climate is considered . Therefore , though the climate , especially adequate rainfall , in the early summer of 2014 is more favorable for mosquito growth and dengue transmission , the lower initial water level of 2014 caused by emptying water containers in late 2013 appears to have reduced the risk to a level even lower than that of 2013 . In other words , the success rate of a single imported case in causing local transmission was not significantly higher in the time window of interest in 2014 and suggests other explanations , such as more imported cases in that time period or by chance alone . The difference between Scenario 2014 and 2014 w/o intervention , as well as between Scenario 2013 and 2013 w/o intervention shows the effectiveness of intervention in reducing dengue transmission probability ( Fig 4B ) . Similar to the three scenarios shown in Fig 4A , the success rates of Scenario 2013 , 2014 w/o intervention , and Scenario 2013 w/o intervention also increase in the early summer until May . However , the duration of the peak differs . The success rate of Scenario 2014 stays at the peak until July , Scenario 2013 until September , while the two scenarios without intervention continue until October . The intervention started on September 27th in 2013 and July 25th in 2014 , so they have no influence on the success rate in the beginning of these years . The success rates are the same for Scenario 2014 and 2014 w/o intervention , as well as for Scenario 2013 and 2013 w/o intervention in the late spring and early summer . However , the success rate drops 20 days before the first intervention , which indicates that the intervention can reduce the transmission probability of a case imported up to 20 days ago . This could possibly due to the death of exposed or infected mosquitoes in the intervention or reduced virus transmission from imported or secondary cases to mosquitoes caused by mosquito population loss . Besides the success rate of one imported case on any given date , the outbreak probability , which is the probability that at least one indigenous case occurs in a particular year , also depends on the number of imported cases . It is clear that the outbreak probability will increase when there are more imported cases , although the increase may be quite small depending on the timing pattern . The outbreak probability can be calculated from the success rate of all imported cases in that year , once the timing of each imported case is known . Fig 5 illustrates the number of identified imported cases into Guangzhou from other countries or Chinese cities in 2013 , 2014 , and the mean , median and range of 2001 to 2014 . Since the detection and reporting rate of imported cases are likely to increase after August , 2014 , when attention began to be paid to the serious dengue outbreak , only data from January to August are shown here . The number of imported cases known to have entered Guangzhou in May and June , 2014 is four times higher than the average or median of that in 2001 to 2014 , which may be the main determinant of the early local dengue virus transmission in 2014 . A different FES can be produced by stochastic effects in each simulation even with the same input . Fig 6 shows the histograms of the FES + 1 under different introduction dates when using the combination of sample set 1 and Scenario 2014 as an example . The FES for an imported case arriving May 11th , 2014 , which is around the estimated timing of the first successful invasion in 2014 , can range from 0 to 132 , 457 with a mean of 22 , 790 and a standard deviation of 25 , 219 . These histograms also show that the FES of successful invasions decreases with the delay of the imported cases . In other words , though the success rate of an early imported case is low , the FES could be extremely high once it succeeds . Average final outbreak size is the mean FES of successful invasions , while average FES is the mean of all simulations . Since the average FES equals the product of the outbreak probability and the final outbreak size , and outbreak probability has been discussed above , here we first look into the average final outbreak size and then the FES . Because all scenarios show the same pattern , and while the results under the last three scenarios are several magnitudes higher than the first three scenarios , to make the figure readable , only the results of the first three scenarios were shown here . Fig 7A shows that the average final outbreak size decreases exponentially with introduction date , and the decrease rate is affected by climate . Scenario 2014 has the highest average final outbreak size , and Scenario 2014 to Avg has the lowest . According to Fig 4A , the mosquito abundance and success rate is higher in Scenario 2014 , so each case can infect others more easily under this scenario , which will then lead to a higher final outbreak size . Unlike the monotonically decreasing average final outbreak size , the average FES peaks in mid-April ( Fig 7B ) , similar to the conclusion from the deterministic model [7] . In addition , the average FES given by the stochastic model is also comparable to that given by the deterministic model . Though the outbreak size is extremely high before April , the success rate is quite low , while after May , though the success rate reaches its peak , the outbreak size drops dramatically . Therefore , the average FES has the highest value in mid-April , and the peak time changes only slightly with climate . Similar to Fig 4A and Fig 7A , Scenario 2014 has the highest average FES .
In [9] , a stochastic model suggested that the slightly warmer summer was adequate to explain the large dengue outbreak in Madeira , 2012 . However , in the Guangzhou outbreak , the favorable climate was only a necessary , but not sufficient condition , because human activities such as intervention and travel also matter . We find the higher number of imported cases in May and June to be the most important determinant of the early outbreak . Although the excessive rainfall in 2014 did increase the successful invasion rate , this effect was cancelled out by the low initial water level due to the interventions in late 2013 . Because of a small outbreak in 2013 , regular interventions were conducted from late September to early November , which markedly reduced the initial water storage in 2014 . The entomological surveillance data shown in Fig 3C supports this interpretation . Specifically , the MOI , which represents the adult mosquito abundance , was lower in 2014 than in 2013 , despite the more favorable climate . The increased number of imported cases in late April to early July was also identified as the driving force of the 2014 Guangzhou dengue outbreak by a deterministic model , which incorporated the number of imported case by a fitting function [44] . The result of a classification tree also suggested the number of imported cases , monthly average BI , and temperature as the most important determinants of the dengue outbreak occurrence in Guangzhou from 2002 to 2013 [45] . As noted in Fig 8A and 8B , dengue endemic countries Thailand , Singapore , Malaysia and Vietnam had the highest tourist exchange with Guangzhou . Therefore , more attention should be paid to identifying imported cases from these countries when they have significant dengue outbreaks , especially in late spring and early summer . Though the successful invasion rate is likely to be low in this period , the final outbreak size can be extremely large once it succeeds ( Fig 7A ) . In all countries , the dengue incidences in 2014 were lower than 2013 , except for Malaysia ( Fig 8C ) . Although the incidences are somewhat comparable in time , it is questionable to compare the incidence of different countries , because the reporting rates may vary significantly between countries . A phylogenetic analysis suggested that the dengue virus isolated in Guangzhou , 2014 was most similar to the isolates in Singapore , Malaysia , Indonesia and Thailand [46 , 47] . The importation index , which incorporates both the incidence in country of origin and the travel volume between the target city and the country of origin [48 , 49] , also shows that Singapore , Thailand and Malaysia had the highest importation indices in 2013 , and the order changed to Malaysia , Singapore and Thailand in 2014 . However , because of the differences in reporting rates between countries , this result is inconclusive . When compared across time , the importation index of 2014 is lower than that of 2013 , but more imported cases were observed in 2014 , especially in May and June ( Fig 5 ) . Considering the epidemic had not started yet at that time , the excessive number of reported imported cases was unlikely to be caused by increasing reporting rate . One possible explanation is spatial heterogeneity . The importation index assumes that human and vector populations are well-mixed , which means that the contact rate between every person and every vector is the same . However , the actual contact rate can be affected by tour route , local transmission hotspots , as well as personal habits , such as whether take measures to avoid mosquito bites . The other explanation is the different reporting rate . For example , if the reporting rate is lower in Malaysia than in other countries , then its actual contribution to the total importation index should be higher , which may make the pattern of the total importation index more similar to that of Malaysia . The first step of a successful invasion is infecting adult mosquitoes ( Event 12 ) , which is described in the model by transition rate bαhvHiAs/N . Both the biting rate b and the transmission probability from human to vector αhv ranges from 0 to 1 , Hi equals to 1 , and the total human population N changed little in the study time period . Hence , the size of the susceptible mosquito population , As , has the greatest influence on this transition rate . As can be seen from Figs 4 , 9A and 9B , it is obvious that the success rate and As have almost the same pattern . Furthermore , we calculated the cross-correlation function ( CCF ) between the difference of success rate and the difference of As . First differences were taken here to remove trend of the time series . The result shows that success rate at day t is positively correlated with As at day t + 10 for all parameter sets under all scenarios . The average result for Scenario 2014 is shown in Fig 9C . Since the average recovery period is assumed to be 6 days in the model [7] , and we use a 10-day time window for the imported case , the 10-day time lag between As and success rate is reasonable . When we explore the individual results for each of the 100 parameter sets , we found that the gap between the success rate under different climate scenarios is more marked when the maximum water level ωmax in that parameter set is larger ( Fig 9D ) . This is because precipitation can affect the carrying capacity through water availability only before ωmax is reached , since water will overflow after reach this threshold . Therefore , if the ωmax is larger , it takes longer to reach this threshold , which will then lead to larger difference in the mosquito abundance and success rate . This phenomenon implies that reducing ωmax by destroying breeding sites can mitigate the effect of climate on dengue transmission dynamics . Regression models are widely used to predict the occurrence of dengue outbreaks , monthly reported cases and the FES [5 , 6 , 45 , 50] . Here we suggest the factors to include in these models according to the results of our stochastic model . Once given the timing of each imported case , the outbreak probability can be calculated from the success rate curve ( Fig 4 ) , which depends mainly on mosquito abundance . Therefore , to predict the occurrence of dengue outbreaks in subtropical areas , variables , such as the number of imported cases and mosquito abundance , should be included . Mosquito abundance is related to temperature , water availability and interventions . Water availability can be estimated from rainfall , evaporation , intervention , and the water limitation parameters ωmax and ωmax . Without all these data , water availability can be indirectly represented by precipitation , and relative humidity . However , precipitation and relative humidity do not reflect the effects of intervention and the relationship between water level and precipitation or relative humidity is non-linear . The water level increases with precipitation in spring and early summer , but when it reaches its threshold ωmax , the water level cannot increase further and remains at this value until evaporation outweighs precipitation or the occurrence of an intervention . In addition , the relationship between water level and mosquito abundance is complicated , which depends on previous water level , the environmental carrying capacity and the previous abundance of aquatic stages . As a result , mosquito indices might still be preferred among all these variables . By including the monthly maximum and minimum temperature , average BI , and number of imported cases , a classification tree can be used to predict the dengue outbreak occurrence in Guangzhou [45] . For tropical areas , since local epidemics continue throughout the entire year , there is no need to predict the outbreak occurrence . The FES depends on both the final outbreak size and outbreak probability . The final outbreak size decreases exponentially as a function of the introduction date , while temperature , water storage , and human interventions can modify the decrease rate through changing the force of infection . Earlier introduction in subtropical areas means a longer period of transmission and more cases , because dengue transmission is terminated by the low temperature every winter . The force of infection can be affected by mosquito abundance , biting rate and virus virulence . Although both the length of transmission and the force of infection can affect the final outbreak size exponentially , the former fluctuates over a much wider range . Sometimes , though the climate is favorable for vector growth , no big outbreak occurs because of the low level or late importation . Thus the imported cases introduce high uncertainty to the prediction models , and a successful model should ideally include the number of imported cases and indigenous cases in the previous month to reduce this uncertainty . Time series regression models with the number of imported and indigenous cases in the previous month , lagged temperature , precipitation and relative humidity can give reasonable estimations of the number of new cases [6 , 50] . Climate conditions in tropical areas can support overwinter local transmission , so the role of imported cases here is not as important as that in subtropical areas . Climate and interventions are the only crucial factors in tropical areas . Prediction models based on only climate can have good prediction power [51–53] . When considering the spatial aspects of the transmission , heterogeneity in the distribution of population , open water , vegetation , and host immunity can also influence the outbreak probability and FES for both subtropical and tropical countries [54 , 55] . Currently , the most common interventions in China are chemical insecticide spraying and environmental management , such as water container emptying . Also , releasing mosquito larvae-eating fish Gambusia affins and wolbachia-infected male mosquitoes are sometimes used in small scale tests . Since the adult abundance and success invasion rate of imported cases increase sharply in April and May ( Fig 4 ) , the interventions should begin before or in this time period . However , the regular interventions in Guangzhou started on September 27th and July 25th in 2013 and 2014 , respectively , both of which were at least one month later than the onset of local epidemics . In addition , interventions out of the transmission season , such as before the next transmission season can also be important , because it can reduce the water storage , increase the time needed to accumulate water in the next transmission season , and postpone the outbreak , which , in turn , can then decrease the final outbreak size significantly . According to result of the PRECIS ( Providing Regional Climate for Impact Study ) regional climate modeling system , the spring and summer rainfall of South China will increase in the future [56] , which may increase the average FES by causing earlier outbreaks . However , destroying mosquito breeding sites can mitigate this detrimental impact of climate change , since it can reduce the maximum water level ωmax and narrow the gaps in the success rate or final outbreak size between different climate scenarios ( Fig 9D ) . Increasing international travel , urbanization , global warming , and changes in precipitation patterns pose higher dengue outbreak risk for Guangzhou and other subtropical non-endemic areas . Moreover , secondary infection can worsen the situation by increasing the incidence of life-threatening dengue hemorrhagic fever and shock syndrome [57] . Under this situation , vector control and management of imported cases at entry points should be implemented strictly , especially around mid-April , to reduce possible health and economic loss caused by dengue or other mosquito-borne diseases , such as yellow fever and Zika . Since the travel volumes between Thailand , Malaysia , Singapore and Guangzhou are the highest , more attentions should be paid to the detection of imported cases when dengue outbreaks occur in these countries , especially at early summer . Besides international imported cases , cases from other Chinese cities should also be considered . Since Guangzhou is normally the first city to have dengue outbreak sin mainland China , it is unusual for Guangzhou to have domestic imported cases in the early phase of an epidemic , although it happens on occasion . For example , the epidemic in Zhongshan started 2 weeks earlier than in Guangzhou [50] . On the other hand , the onset of an epidemic in Guangzhou signals that surrounding cities of the need to pay attention to case detection and vector control immediately . For example , from Figs 4B and 9C , it can be seen that , after detection , the model suggests that interventions should be conducted around the residences of the reported cases whose symptoms began less than 10 days ago , even if they have recovered . The model proposed here represents the transmission of only one serotype DENV-1 , since 5 , 947 out of 6 , 024 cases ( 98 . 7% ) tested in 2014 were infected by DENV-1 , and only 74 and 3 were infected by DENV-2 and -3 , respectively [48] . When adapted for use in other subtropical areas , the single serotype assumption may or may not be appropriate . Secondly , we assumed that dengue virus virulence and reporting rate were the same for 2013 and 2014 , which led to similar initial exponential growth rates for these two years . However , from the daily reported new case data , the growth rate of 2014 was slightly higher than that of 2013 ( Fig 3 ) , the reasons for which deserve further investigation . Thirdly , the estimated success rates illustrated in Fig 4 are estimates of the upper bound on what might be expected in practice , because it estimates the success rates for imported cases who spend the whole viremic period in Guangzhou . However , an imported case can enter Guangzhou at the middle or even the last day of the viremic period and thereby have less time to infect Ae . albopictus and result in a lower successful invasion rate . In this model , we also assumed that humans and the mosquitoes are well-mixed , which means the contact rate is the same for every host and vector . However , studies of Ae . albopictus indicate that their flying range is as small as 300 meters , and that they normally they stay near their breeding sites for their whole life [58] . Therefore , a spatially-explicit model is needed to better simulate the spatial distribution of dengue cases and study the risk factors related to the observed distribution patterns in the future . | An unprecedented dengue outbreak occurred in Guangzhou , 2014 , with 38 , 036 reported cases in contrast to 73 , 179 cases in all of mainland China from 1990 to 2015 . In an earlier analysis using a deterministic model , we concluded the early timing of local transmission to be the most important determinant of this outbreak . Here we use a stochastic model to explore the reasons why the outbreak happened earlier in 2014 . Our results identified the higher number of imported cases in May and June to be the most probable explanation . Based on the investigation of the determinants of success rate and final epidemic size , this work provides suggestions for reducing dengue outbreak potential and epidemic size in the future . More attention should be paid to imported case detection and vector control measures in early summer , because this is the time when successful invasion can result in high incidence of infection and the success rate of each imported case begins to rise . Destroying mosquito breeding sites can reduce the maximum water level of the system and attenuate the role played by climate . In addition , interventions within 10 days after the introduction of imported cases is still effective in preventing further transmission . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"death",
"rates",
"invertebrates",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"demography",
"china",
"atmospheric",
"science",
"pathogens",
"geographical",
"locations",
"microbiology",
"animals",
"viruses",
"seasons",
"summer",
"developmental",
"biology",
"rna",
"viruses",
"insect",
"vectors",
"infectious",
"diseases",
"medical",
"microbiology",
"microbial",
"pathogens",
"life",
"cycles",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"asia",
"flaviviruses",
"meteorology",
"earth",
"sciences",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"malaysia",
"larvae",
"organisms"
] | 2017 | The interplay of climate, intervention and imported cases as determinants of the 2014 dengue outbreak in Guangzhou |
Cell type-specific modifications of conventional endosomal trafficking pathways lead to the formation of lysosome-related organelles ( LROs ) . C . elegans gut granules are intestinally restricted LROs that coexist with conventional degradative lysosomes . The formation of gut granules requires the Rab32 family member GLO-1 . We show that the loss of glo-1 leads to the mistrafficking of gut granule proteins but does not significantly alter conventional endolysosome biogenesis . GLO-3 directly binds to CCZ-1 and they both function to promote the gut granule association of GLO-1 , strongly suggesting that together , GLO-3 and CCZ-1 activate GLO-1 . We found that a point mutation in GLO-1 predicted to spontaneously activate , and function independently of it guanine nucleotide exchange factor ( GEF ) , localizes to gut granules and partially restores gut granule protein localization in ccz-1 ( - ) and glo-3 ( - ) mutants . CCZ-1 forms a heterodimeric complex with SAND-1 ( MON1 ) , which does not function in gut granule formation , to activate RAB-7 in trafficking pathways to conventional lysosomes . Therefore , our data suggest a model whereby the function of a Rab GEF can be altered by subunit exchange . glo-3 ( - ) mutants , which retain low levels of GLO-3 activity , generate gut granules that lack GLO-1 and improperly accumulate RAB-7 in a SAND-1 dependent process . We show that GLO-1 and GLO-3 restrict the distribution of RAB-7 to conventional endolysosomes , providing insights into the segregation of pathways leading to conventional lysosomes and LROs .
Caenorhabditis elegans gut granules are lysosome related organelles ( LROs ) [1] , cell type-restricted compartments with diverse functions that share characteristics with conventional lysosomes [2] . These conspicuous intestine-specific compartments contain birefringent and autofluorescent material [3–5] . Gut granules function in lipid transport [6] , metabolism [7 , 8] , and signaling [9] , as well as the storage and detoxification of metals and xenobiotics [10–12] . Gut granules coexist with conventional lysosomes and are not the major degradative compartments within C . elegans intestinal cells [1 , 5 , 13 , 14] . LRO biogenesis is mediated by evolutionarily conserved pathways that divert cargo away from conventional endosomes toward LROs [15 , 16] . Defects in these pathways cause Hermansky-Pudlak syndrome , a human condition characterized by a lack of dense granules and malformed melanosomes , LROs found within platelets and melanocytes , respectively [17] . Screens for C . elegans mutants that disrupt gut granule biogenesis have identified conserved factors that promote LRO biogenesis . Similar to many mammalian LROs , gut granule formation requires the BLOC-1 [18] , HOPS [19] , and AP-3 complexes [1] , LYST [20] , and the Rab32 family member GLO-1 [1] . Rab32 family members have a conserved function in LRO biogenesis and are one of 20 Rabs hypothesized to have been present in the last eukaryotic common ancestor [1 , 21–28] . The reversible association of Rabs with membranes , where they regulate key cargo trafficking events , is in large part mediated by guanine nucleotide exchange factors ( GEFs ) [29–32] . In mammals , the BLOC-3 GEF activates and localizes the Rab32 family members Rab32 and Rab38 [33] . BLOC-3 is a heterodimeric complex composed of two proteins , HPS1 and HPS4 [34–36] . Despite the presence of a Rab32 family member , HPS1 and HPS4 are not obviously conserved in C . elegans [37] , raising the question of how GLO-1 is activated and localized . HPS1 and HPS4 have weak homology to CCZ-1 and SAND-1 ( MON1 ) , respectively [33 , 38 , 39] , which form a heterodimeric GEF activating Rab7 in the conventional endolysosomal trafficking pathway [40–43] . Rab7 and Rab32 family members function in endosomal trafficking pathways , are closely related , and might share regulatory factors or effectors [22] . We recently found that ccz-1 ( - ) mutants lack gut granules , whereas rab-7 ( - ) mutants have only a minor defect in gut granule protein localization , and sand-1 is dispensable for gut granule biogenesis [19] . These phenotypes indicate an unexpected role for CCZ-1 in LRO biogenesis that does not involve its known interactions with SAND-1 ( MON1 ) or its regulation of RAB-7 . Given the Rab GEF activity of CCZ-1 and the homology of Rab32 family members with Rab7 , CCZ-1 might function with a different protein to activate and localize GLO-1 . Our recent work identified weak homology between GLO-3 and HPS1 and placed GLO-3 , like CCZ-1 , function upstream of GLO-1 activation , providing an initial indication that this interacting factor might be GLO-3 [19] . Here we present the results of studies analyzing the physical association of GLO-3 and CCZ-1 and the functional relationships between CCZ-1 and GLO-3 in GLO-1 localization and gut granule and conventional endolysosome biogenesis . Our results show that CCZ-1 and GLO-3 function to localize GLO-1 to LRO membranes and promote gut granule formation . However , GLO-3 and GLO-1 do not function in endolysosome biogenesis like CCZ-1 and instead act to restrict RAB-7 to conventional endolysosomes .
To determine if GLO-3 can physically interact with CCZ-1 , we screened for interactions between the two full-length proteins using the yeast two-hybrid system . When used as bait , CCZ-1 interacted with the GLO-3 prey , promoting the expression of both LEU2 and LacZ reporters ( Fig 1A ) . We confirmed the interaction between GLO-3 and CCZ-1 by expressing the proteins in E . coli and using a glutathione S-transferase ( GST ) -pull down assay . Full length GST-GLO-3 pulled down both full length CCZ-1 and the CCZ-1 ( 1–200 ) amino terminal region ( Fig 1B ) . GLO-3 ( 1–219 ) was able to pull down CCZ-1 ( 1–200 ) , albeit not as strongly as full length GLO-3 ( Fig 1B ) . The GST moiety attached to GLO-3 did not significantly interact with either form of CCZ-1 ( Fig 1B ) . Taken together , these results show that GLO-3 and CCZ-1 directly interact and that the amino-terminal domain of CCZ-1 , which contains a longin domain [40] , acts as a binding interface between the two proteins . We analyzed whether the effects of mutations in glo-1 and glo-3 on gut granule and endolysosome biogenesis resembled the loss of ccz-1 function . The glo-1 ( zu437 ) allele used in our studies completely lacks GLO-1 activity [1] . We have isolated a glo-3 allelic series spanning four phenotypic classes ( I-IV ) [44] , ranging from the least severe , a newly identified class IV glo-3 ( gk582755 ) missense allele GLO-3 ( N279K ) that causes a moderate reduction in gut granule numbers , to the strongest class I alleles , represented by glo-3 ( kx94 ) , which lack birefringent gut granules in embryos but retain a few autofluorescent gut granules in adults ( Tables 1 and 2 ) . Nearly all of the glo-3 alleles are premature stop codons and their location in glo-3 does not correlate with their phenotypic severity [44] . We note that the two class I alleles cause amber stop codons , while the weaker class II and class III alleles cause ochre or opal stop codons [44] . In C . elegans , opal and ochre stop codons can occasionally be read through during translation , while amber stop codons cannot [45] , suggesting that the class II and III alleles produce differing amounts of full length GLO-3 . CRISPR-Cas9 was used to delete the entire glo-3 coding sequence and the resulting glo-3 ( syb272 ) allele was phenotypically indistinguishable from class I mutants ( S1 Fig and Tables 1 and 2 ) , strongly suggesting that the glo-3 ( kx94 ) allele used in this work represents a null allele . In prior work , we showed that disrupting the function of glo-1 or glo-3 leads to significant reductions in autofluorescent and birefringent gut granules ( Tables 1 and 2 ) [1 , 19] . To determine if this is associated with defects in protein localization in glo-1 ( - ) and glo-3 ( - ) mutants , we analyzed the steady state distribution of the gut granule transmembrane proteins CDF-2 , which functions as a Zn transporter [46] , the ABC transporter PGP-2 [47] , and the Lamp1 homologue LMP-1 [48] . CDF-2 and PGP-2 are restricted to gut granules , while LMP-1 is associated with both gut granules and conventional lysosomes [13] . To minimize indirect effects , we carried out our analyses in embryonic intestinal cells soon after gut granules are first generated [3] . In glo-1 ( zu437 ) and glo-3 ( kx94 ) mutants , the distribution of CDF-2::GFP and LMP-1 was dramatically altered ( Fig 2B and 2C ) . Strikingly , both were mislocalized to the intestinal cell membrane ( Fig 2B and 2C ) . While the extensive colocalization of CDF-2::GFP and LMP-1 was not disrupted in these mutants , they were not located on organelles that resembled gut granules ( Fig 2B , 2C and 2G ) . Instead , the morphology and position of the compartments containing these markers were similar to conventional lysosomes . To examine whether CDF-2::GFP was being mislocalized to lysosomes , we analyzed the distribution of CDF-2::GFP relative to an mCherry tagged form of GBA-3 , a glucosylceramidase localized to degradative lysosomes that when disrupted in humans causes Gaucher disease [13 , 49 , 50] . We found that CDF-2::GFP was mislocalized to conventional lysosomes in glo-1 ( - ) and glo-3 ( - ) mutants ( Fig 2D , 2F and 2I ) , explaining why the colocalization of CDF-2::GFP and LMP-1 was not altered in these mutants . PGP-2 was lacking in both glo-1 ( - ) and glo-3 ( - ) mutants ( Fig 2B , 2C and 2H ) , possibly due to its lysosomal mistargeting and degradation . These results indicate that GLO-1 and GLO-3 both play essential roles in the routing of proteins to gut granules , similar to what we have previously seen for CCZ-1 [19] . We next addressed whether GLO-1 and GLO-3 function in the formation of lysosomes . Gut granules and conventional lysosomes are distinct and co-exist in embryonic intestinal cells [13] . Some factors that mediate gut granule biogenesis , including CCZ-1 , also function in lysosome biogenesis and disrupting their activity causes a significant increase in the number of endolysosomes marked by LMP-1::GFP [19] . LMP-1::GFP , like endogenous LMP-1 , is localized to conventional lysosomes , however the addition of GFP to its cytoplasmic carboxyl-terminus disrupts it’s sorting , causing its loss from gut granules and enrichment at the cell membrane [13] . In glo-1 ( - ) mutants there was a slight decrease in the number of LMP-1::GFP compartments and glo-3 ( - ) mutants were unchanged relative to wild type ( Fig 3A and 3B ) , which is very different than what we see in ccz-1 ( - ) mutants , where the number is increased two-fold [19] . RAB-7 dynamically localizes to endolysosomes as they mature and defective trafficking to conventional lysosomes significantly alters the colocalization of RAB-7 and LMP-1::GFP [19 , 51] . Approximately 50% of LMP-1::GFP compartments were labeled by RAB-7 in wild type , a level of colocalization that was not markedly different in glo-1 ( - ) and glo-3 ( - ) mutants ( Fig 3C–3F ) . In C . elegans coelomocytes , decreased trafficking to lysosomes suppresses the enlargement of lysosomes caused by mutations in CUP-5 [52] . CUP-5 is orthologous to human TRPML1 [53 , 54] , which is mutated in type IV mucolipidosis , and mediates the formation of lysosomes from endosomal-lysosomal hybrid compartments [55] . LMP-1 containing endolysosomes were enlarged in cup-5 ( zu223 ) embryonic intestinal cells ( Fig 3G , 3H and 3L ) . Moreover , rab-7 ( RNAi ) disrupts trafficking to lysosomes and reduced cup-5 ( - ) endolysosomes back to wild-type size , validating the assay in this cell type ( Fig 3I and 3L ) . In contrast , endolysosomes in cup-5 ( - ) ; glo-1 ( - ) and cup-5 ( - ) ; glo-3 ( - ) double mutants remained significantly enlarged , albeit not quite as large as in cup-5 ( - ) ( Fig 3J–3L ) . Together these observations point to GLO-1 and GLO-3 having an essential role in localizing gut granule proteins and a minor , if any , role in trafficking to conventional lysosomes . In glo-1 ( - ) and glo-3 ( - ) mutants , the mislocalization of the gut granule protein CDF-2::GFP to conventional lysosomes led to a high level of colocalization between CDF-2::GFP and LMP-1 , which marks lysosomes ( Figs 2 , 4D , 4G and 4L ) . In contrast , only 50% of CDF-2::GFP compartments were marked by LMP-1 in ccz-1 ( - ) mutants ( Fig 4C and 4L ) , revealing a significant phenotypic difference in gut granule protein localization between ccz-1 ( - ) and the glo-1 ( - ) and glo-3 ( - ) mutants . In double mutants , the colocalization phenotype of ccz-1 ( - ) was epistatic to glo-1 ( - ) and glo-3 ( - ) ( Fig 4F , 4I and 4L ) . While it is currently unclear whether ccz-1 ( - ) impacts the colocalization of CDF-2::GFP and LMP-1 by mislocalizing CDF-2::GFP , LMP-1 , or both proteins , the data indicate that CCZ-1 has functions in protein localization distinct from GLO-1 and GLO-3 . We investigated whether sand-1 ( - ) mutants disrupt the colocalization of CDF-2::GFP and LMP-1 similarly to ccz-1 ( - ) , due to the well-established role of the C . elegans CCZ-1/SAND-1 ( MON1 ) complex in the biogenesis of late endosomes and trafficking to conventional lysosomes [42 , 56 , 57] . In sand-1 ( - ) single mutants , the colocalization of CDF-2::GFP with LMP-1 was not obviously different than wild type ( Fig 4B and 4L ) . However , gut granule biogenesis is not disrupted in sand-1 ( - ) mutants like it is in ccz-1 ( - ) mutants [19] . When sand-1 ( - ) was combined with glo-1 ( - ) or glo-3 ( - ) mutants the colocalization of CDF-2::GFP and LMP-1 was indistinguishable from that of ccz-1 ( - ) single mutants ( Fig 4B , 4E , 4H and 4L ) . This effect is consistent with SAND-1 functioning only in localizing proteins to endolysosomes , GLO-1 and GLO-3 functioning only in gut granule protein localization , and CCZ-1 functioning in both processes . In support of this interpretation , disrupting the function of apt-7 , which encodes a subunit of the AP-3 complex that functions in trafficking to both LROs and conventional lysosomes [58] , altered the colocalization of CDF-2::GFP with LMP-1 similar to ccz-1 ( - ) , and the addition of sand-1 ( - ) did not modify its effects ( Fig 4J–4L ) . GLO-1 is a Rab32 family member and the conservation of G-motifs suggests that it functions as a GTPase [21 , 22] . To determine if GTP binding is important for the activity of GLO-1 in vivo we expressed GLO-1 point mutants in glo-1 ( zu437 ) animals . Whereas GFP tagged GLO-1 ( + ) restored gut granules in glo-1 ( zu437 ) ( Fig 5A–5C ) , expression of GFP::GLO-1 ( T25N ) , which is predicted to disrupt GTP but not GDP binding [59] , did not rescue the loss of autofluorescent , birefringent , and PGP-2 marked gut granules ( Fig 5E and Tables 1 and 2 ) . Expression of GFP::GLO-1 ( Q71L ) , which is predicted to lack GTP hydrolysis and maintain an active GTP-bound conformation [59] , was able to functionally replace GLO-1 ( + ) ( Fig 5D and Tables 1 and 2 ) . Neither GLO-1 point mutant dominantly disrupted gut granule biogenesis ( Tables 1 and 2 ) . The different mutant effects suggest that the GTP bound form of GLO-1 is active in promoting gut granule biogenesis . Mutations in the Rab G4 motif can weaken its affinity for guanine nucleotides leading to increased rates of nucleotide exchange that can bypass the requirement of a Rab for its corresponding GEF [41 , 60 , 61] . GLO-1 ( D132A ) and GLO-1 ( I133F ) G4 motif mutations restore autofluorescent compartments in ccz-1 ( - ) and glo-3 ( - ) mutant adults [19] , suggesting that spontaneous nucleotide exchange bypasses the requirement for CCZ-1 or GLO-3 in gut granule biogenesis . The vha-6 promoter used in our prior study initiates expression late in embryogenesis [62 , 63] . To assess the effects of the GLO-1 G4 mutants at a stage when we can rigorously assess the biogenesis of gut granules using multiple organelle markers , we placed the point mutants under control of the glo-1 promoter , which leads to earlier intestinal expression . When introduced into glo-1 ( zu437 ) mutants , both GLO-1 G4 mutants restored birefringent compartments in embryonic intestinal cells and autofluorescent intestinal organelles in adults ( Fig 6A–6D and Tables 1 and 2 ) . We focused our analysis on GLO-1 ( D132A ) as it had the strongest rescuing activity ( Tables 1 and 2 ) . GLO-1 ( D132A ) restored birefringent and autofluorescent granules in glo-3 ( - ) and ccz-1 ( - ) mutants , but did not suppress the loss of these organelles in AP-3 , BLOC-1 , or HOPS mutants ( Fig 6F and 6J and Tables 1 and 2 ) . Expression of GLO-1 ( + ) only restored birefringent and autofluorescent organelles in glo-1 ( - ) mutants ( Fig 6C , 6E and 6I and Tables 1 and 2 ) . Therefore , GLO-3 and CCZ-1 likely function upstream of GLO-1 in the formation of gut granules . In the ccz-1 ( - ) and glo-3 ( - ) mutant strains where gut granule biogenesis was rescued by GLO-1 ( D132A ) , endogenous wild-type glo-1 ( + ) is also present . GLO-1 ( D132A ) was introduced into a glo-1 ( - ) glo-3 ( - ) double mutant where it fully restored autofluorescent and birefringent compartments ( Fig 6H and Tables 1 and 2 ) . This result indicates that the restoration of gut granules in glo-3 ( - ) is mediated by GLO-1 ( D132A ) and not by endogenous glo-1 ( + ) . We investigated whether GLO-1 ( D132A ) restored gut granule protein localization in glo-1 ( - ) , glo-3 ( - ) , and ccz-1 ( - ) mutants . For these experiments , we quantified the colocalization of proteins using SQUASSH image analysis software that deconvolves , segments , and calculates the overlapping area between two fluorescence signals in three dimensions [64] . This software enabled a comprehensive , high throughput , and quantitative approach for identifying and measuring the area of overlap between two different organelle markers within the entire embryonic intestine . The output Csize ( marker 1marker2/marker 1 ) describes the proportion of marker 1’s area that contains marker 2 . PGP-2 labeled organelles were lacking in glo-1 ( - ) , glo-3 ( - ) , and ccz-1 ( - ) mutant embryos ( Fig 2B and 2C ) . Expression of GLO-1 ( + ) or GLO-1 ( D132A ) in glo-1 ( - ) embryos restored gut granules containing PGP-2 and LMP-1 ( Fig 7B and 7E and S2 Fig ) . We note that GLO-1 ( D132A ) did not always support the high level of colocalization seen between these markers when GLO-1 ( + ) was expressed ( Fig 7B and 7E and S2 Fig ) . Consistent with our observations of other gut granule markers ( Tables 1 and 2 ) , expression of GLO-1 ( + ) in glo-3 ( - ) or ccz-1 ( - ) mutants did not restore PGP-2 compartments ( S2 Fig ) . In contrast , GLO-1 ( D132A ) robustly supported the formation of PGP-2 marked organelles in both mutants ( Fig 7C , 7D and 7E ) . The rescuing activity of GLO-1 ( D132A ) was not complete however , as these compartments lacked LMP-1 ( Fig 7C , 7D and 7E ) . Therefore , while GLO-1 ( D132A ) can substitute for much of the activity of glo-3 ( + ) and ccz-1 ( + ) in the biogenesis of gut granules and the localization of PGP-2 , the pathway that directs LMP-1 to gut granules is distinct and more sensitive to alterations in GLO-1 activity . Inactive Rabs are GDP-bound and can be extracted from organelle membranes into the cytosol by Rab GDI , while activated , GTP-bound Rabs are resistant to extraction and membrane localized [29 , 31] . The restoration of gut granules in glo-3 ( - ) and ccz-1 ( - ) mutants by GLO-1 ( D132A ) is consistent with GLO-3 and CCZ-1 functioning upstream of GLO-1 activation . We therefore examined whether GFP::GLO-1 was cytoplasmic and diffusely localized or enriched on organelles . For these experiments , we examined living embryos by imaging GFP::GLO-1 and performed line intensity scans to compare the organelle and cytoplasmic signals . In wild type , GFP::GLO-1 was localized to gut granules and relatively little signal was found in the cytoplasm ( Fig 8A and 8I ) . When GFP::GLO-1 was identically imaged , both glo-3 ( - ) and ccz-1 ( - ) mutants displayed a diffuse signal and GFP::GLO-1 was lacking from prominent puncta ( Fig 8G and 8H ) . Line scans through sites of GFP::GLO-1 enrichment showed that strongly labeled GFP::GLO-1 structures were missing from both mutants ( Fig 8G , 8H and 8I ) . glo-3 ( - ) and ccz-1 ( - ) mutants lack gut granules ( Figs 2 , 6 and S2 ) , which could explain the altered distribution of GFP::GLO-1 . However , an apt-7 ( - ) mutant that disrupts the AP-3 complex [1] , a vps-18 ( - ) mutant that disrupts the HOPS complex [19] , a snpn-1 ( - ) mutant that disrupts the BLOC-1 complex [18] , and a glo-4 ( - ) mutant [1] , all lack or have few gut granules and they all displayed a very different GFP::GLO-1 pattern . In these mutants , GFP::GLO-1 was enriched on small puncta and was not diffusely localized ( Fig 8B–8E and 8I ) , indicating that the loss of gut granules per se does not lead to the diffuse localization of GFP::GLO-1 in the glo-3 ( - ) and ccz-1 ( - ) strains . CCZ-1 functions with SAND-1 ( MON1 ) as a RAB-7 GEF promoting early to late endosome conversion in the conventional lysosomal trafficking pathway [51] . GFP::GLO-1 localization in sand-1 ( - ) mutants resembled wild type , albeit with less brightly labeled organelles ( Fig 8F and 8I ) , suggesting that disrupting endosome maturation does not cause the diffuse pattern of GFP::GLO-1 in glo-3 ( - ) and ccz-1 ( - ) mutants . The class III glo-3 ( kx38 ) allele , which generates gut granules marked by PGP-2 ( see next section ) , was used to address whether GLO-1 was localized to gut granules when glo-3 function was partially , but not completely , disrupted . In fixed wild-type embryos , GFP::GLO-1 was associated with gut granules , whereas it was lacking from PGP-2 marked gut granules in the glo-3 ( kx38 ) mutant ( Fig 7F and S2 ) . To test whether the ectopically expressed and epitope tagged GFP::GLO-1 behaves similar to endogenous GLO-1 , we stained wild-type and glo-3 ( kx38 ) embryos with anti-GLO-1 antibodies . In wild type , GLO-1 localized to gut granules marked by CDF-2::GFP ( Fig 8J and 8L ) . In contrast , GLO-1 was lacking from gut granules in glo-3 ( kx38 ) mutants ( Fig 8K and 8L ) . Taken together these results show that GLO-3 and CCZ-1 promote the association of GLO-1 with gut granules . In both wild-type and glo-1 ( - ) embryos , GFP::GLO-1 ( D132A ) colocalized with PGP-2 similarly to GFP::GLO-1 ( + ) ( Figs 7A , 7B and 7F and S2 ) , demonstrating that the GLO-1 ( D132A ) fast exchange mutant can properly associate with gut granules . Since both glo-3 ( - ) and ccz-1 ( - ) mutants mislocalized GFP::GLO-1 ( + ) ( Fig 7G–7I ) , we addressed whether GFP::GLO-1 ( D132A ) similarly required glo-3 and ccz-1 to localize to gut granules . In the class III glo-3 ( kx38 ) mutant , GFP::GLO-1 ( D132A ) properly localized to PGP-2 containing compartments at levels similar to when it is expressed in glo-1 ( - ) ( Figs 7F and S2 ) . In the glo-3 ( kx94 ) mutant GFP::GLO-1 ( D132A ) often localized to gut granules ( Fig 7C and 7F ) and GFP::GLO-1 ( D132A ) occasionally associated with gut granules in ccz-1 ( - ) mutants , colocalizing with gut granules at lower levels than seen in glo-1 ( - ) and the glo-3 ( - ) mutants ( Fig 7D and 7F ) . The GLO-3 and CCZ-1 independent localization of GFP::GLO-1 ( D132A ) indicate that these proteins are not absolutely required to localize GLO-1 when its spontaneous exchange activity is increased . All of the glo-1 alleles we have isolated and the glo-3 class I null alleles , including glo-3 ( kx94 ) , completely disrupt embryonic gut granule biogenesis ( Figs 2 and 6 and Table 1 ) [1 , 44] , making it difficult to determine how GLO-1 and GLO-3 function in the pathways generating gut granules . We have isolated a large number of glo-3 ( - ) mutants with varying levels of glo-3 ( + ) activity and a range of phenotypic effects ( Tables 1 and 2 ) [44] . This allelic series can reveal phenotypes that result from partial glo-3 ( + ) activity and point to specific functions for GLO-3 and the GLO-1 Rab it regulates in gut granule biogenesis . We analyzed the class III allele glo-3 ( kx38 ) [44] , and found that glo-3 ( kx38 ) mutants generate organelles that have many gut granule characteristics . First , birefringent granules are always generated in glo-3 ( kx38 ) embryos , albeit at reduced numbers ( Table 1 ) . Second , the formation of birefringent granules in glo-3 ( kx38 ) embryos required the activity of AP-3 ( apt-7 ) and BLOC-1 ( glo-2 and snpn-1 ) subunits , as well as other genes ( glo-4 and pgp-2 ) necessary for gut granule biogenesis ( Table 3 ) . Third , PGP-2 marked organelles were present in glo-3 ( kx38 ) mutants and their formation required the same genes ( Figs 9B and 9D and S3 ) . Fourth , compartments containing both of the gut granule proteins PGP-2 and CDF-2::GFP were present in glo-3 ( kx38 ) mutants ( Fig 9B and 9E ) . Finally , the number of organelles marked by PGP-2 was similar to the numbers of birefringent organelles generated in this mutant ( Fig 9B and 9G and Table 3 ) . We examined whether the gut granule-like organelles in glo-3 ( kx38 ) had any characteristics of endolysosomes . LMP-1::GFP and two different lysosomal hydrolases , GBA-3::mCherry and CPR-6::mCherry , which is a tagged cathepsin B peptidase [13 , 65 , 66] , were not mislocalized to these compartments ( S4 Fig ) . In addition , birefringent material and CDF-2::GFP remained associated with PGP-2 marked organelles when glo-3 ( kx38 ) was combined with cup-5 ( - ) , which disrupts conventional endolysosomal trafficking and inhibits lysosome formation [55] ( S3 Fig and Table 3 ) . Taken together , these observations indicate that gut granules are generated in glo-3 ( kx38 ) mutants . While gut granules are present in glo-3 ( kx38 ) embryos , our work shows that both GFP::GLO-1 and endogenous GLO-1 are lost from these compartments ( Figs 7F and 8K and S2 ) . Therefore an analysis of this mutant can reveal effects on LRO biogenesis when glo-3 ( + ) activity is reduced and GLO-1 is lacking from gut granules . The most obvious effect of glo-3 ( kx38 ) is on gut granule number and size; compared to wild type , the number of gut granules marked by PGP-2 was reduced by more than 80% and the average gut granule diameter was 55% larger ( Fig 9F and 9G ) . To determine if glo-3 ( kx38 ) disrupts protein trafficking , we analyzed the localization of gut granule markers in glo-3 ( kx38 ) mutants . LMP-1 is localized to both gut granules and conventional lysosomes [13] , and in glo-3 ( kx38 ) mutant embryos LMP-1 was mislocalized to the plasma membrane and lacking or only weakly associated with gut granules ( Fig 9C–9E ) . CDF-2::GFP remained associated with gut granules in glo-3 ( kx38 ) mutants ( Fig 9B and 9E ) . However , CDF-2::GFP was mislocalized to the plasma membrane and what are likely conventional lysosomes based upon their morphology , location , and enlargement in cup-5 ( - ) mutants ( Figs 9B and S3 ) . These analyses show that the localization of CDF-2::GFP and LMP-1 are sensitive to reduction in glo-3 activity , while PGP-2 appears to be unaffected . Notably , the presence of gut granules in glo-3 ( kx38 ) mutants indicates that the enrichment of GLO-1 on gut granules is not necessary for their biogenesis . Due to the ability of small GTPases that direct intracellular trafficking to regulate each other’s localization [31 , 67 , 68] , we investigated whether the reduction in glo-3 activity and loss of GLO-1 from gut granules in glo-3 ( kx38 ) mutants had any effects on the gut granule localization of other Rab and Arf GTPases . The early endosomal GFP::RAB-5 , apical recycling endosomal GFP::RAB-11 . 1 , basolateral recycling endosomal GFP::RAB-10 , and lysosomal ARL-8::GFP were not mislocalized to gut granules in glo-3 ( kx38 ) mutants ( S4 Fig ) . In contrast , a significant fraction of gut granules in glo-3 ( kx38 ) , but not wild-type embryos , accumulated the late endosomal GFP::RAB-7 ( S4 Fig ) . Confirming the result with the ectopically expressed and tagged protein , endogenous RAB-7 similarly mislocalized to gut granules in glo-3 ( kx38 ) mutants ( Fig 10C , 10E , 10J and 10M ) . Gut granules in class IV glo-3 ( - ) mutants accumulated RAB-7 as well ( S5 Fig ) . The activation and localization of RAB-7 to endosomes is mediated by CCZ-1/SAND-1 ( MON1 ) [42 , 51 , 56] . RAB-7 was lacking from sand-1 ( - ) ; glo-3 ( kx38 ) gut granules ( Fig 10D and 10E ) , consistent with SAND-1 promoting the association of RAB-7 with gut granules in glo-3 ( kx38 ) mutants . We addressed whether RAB-7 has a role in the formation of gut granules in glo-3 ( kx38 ) mutants . Suggesting that it does not , we found that the formation of birefringent gut granules was not disrupted in glo-3 ( kx38 ) ; rab-7 ( ok 511 ) double mutants ( Table 3 ) . sand-1 ( - ) mutations disrupt RAB-7 activity and localization [42] , and when combined with glo-3 ( kx38 ) did not alter the formation of birefringent gut granules ( Table 3 ) , the number of PGP-2 marked gut granules ( Fig 9G ) , or the colocalization of the gut granule proteins PGP-2 and CDF-2::GFP ( S3 Fig ) . Together , these results strongly suggest that the presence of RAB-7 on gut granules in glo-3 ( kx38 ) does not substantially impact their biogenesis . rab-7 ( - ) mutants generate gut granules whose morphology , number , and protein composition resemble the gut granules in glo-3 ( kx38 ) mutants [19] , suggesting that RAB-7 might play a role in recruiting GLO-1 . However , we found that GLO-1 was present on gut granules in rab-7 ( ok511 ) mutants ( Fig 10F–10H ) , indicating RAB-7 is not necessary for the recruitment of GLO-1 to gut granules . To further investigate the functional relationships between GLO-1 and RAB-7 we ectopically expressed GFP::GLO-1 ( + ) and GFP::GLO-1 ( D132A ) in glo-3 ( kx38 ) mutants and found that both led to a significant decrease in the association of RAB-7 with gut granules ( Fig 10I–10M ) . These results point to a role for GLO-1 in preventing the association of RAB-7 with gut granules and suggest that the association of RAB-7 with gut granules in glo-3 ( kx38 ) mutants could result from the loss of GLO-1 from these organelles . One mechanism by which GLO-1/GLO-3 could restrict RAB-7 from gut granules would be through the recruitment and/or activation of the RAB-7 GTPase activating protein ( GAP ) . Currently , the RAB-7 GAP is not known in C . elegans , however the RAB-5 GAP TBC-2 has RAB-7 GAP activity in vitro , and genetic studies are consistent with it functioning as a RAB-7 GAP [69–71] . However , tbc-2 ( - ) mutants did not lead to the mislocalization of RAB-7 to gut granules or obvious defects in gut granule protein trafficking ( S6 Fig ) .
GLO-1 and related Rab32/38 proteins were initially identified due to their role in the biogenesis of LROs in mammals [24 , 27] , Drosophila melanogaster [25] , and C . elegans [1] . More recently Rab32 family members have been implicated in autophagy [72 , 73] , phagocytosis of bacterial pathogens [74–77] , and endocytosis and proteolytic degradation [78] . In C . elegans , glo-1 ( - ) early embryos are defective in the autophagic degradation of paternal mitochondria [79–81] . In the nervous system , glo-1 ( - ) adults show decreased numbers of RAB-7 labeled compartments [82] , altered necrosis [83] , and defects in synapse formation and neuronal morphology [84] . All of these more recently identified roles for Rab32 family members , including GLO-1 , could result from functions in the conventional endolysosomal pathway . In fact , many of the factors originally characterized as having a role in LRO biogenesis are now known to support conventional endolysosomal trafficking [1 , 19 , 85–89] . Notably , we did not detect a significant role for GLO-1 in the transport of cargo through conventional endolysosomes and instead show that GLO-1 functions to direct protein cargo away from this pathway toward gut granules ( Figs 2 and 3 ) . Our data support an LRO restricted role for GLO-1 in intestinal cells and we suggest that processes mediated by gut granules impact developmental and physiological processes outside the intestine or that other C . elegans cell types possess LROs , with different functions than gut granules , whose formation requires GLO-1 . Similar to other Rabs [29 , 31] , we show that the GTP bound form of GLO-1 is active in gut granule formation ( Fig 5 ) . Following GTP hydrolysis , most Rabs will remain in the inactive form due to their low intrinsic rate of exchange of GDP for GTP [90] . In mammals , Rab32/38 guanine nucleotide exchange is catalyzed by BLOC-3 , a heterodimeric complex composed of HPS1 and HPS4 [33] . BLOC-3 subunits show sequence and functional homology with CCZ-1/SAND-1 ( MON1 ) , which function as a heterodimeric GEF for RAB-7 [33 , 38 , 39 , 41 , 42 , 91 , 92] . In addition to interacting with SAND-1 ( MON1 ) , we find that CCZ-1 can directly bind to GLO-3 ( Fig 1 ) . We have previously shown that CCZ-1 , but not SAND-1 or RAB-7 , is required for gut granule biogenesis , and that a point mutation in GLO-1 predicted to increase the rate of spontaneous guanine nucleotide exchange restores of autofluorescent organelles in ccz-1 ( - ) and glo-3 ( - ) mutant adults [19] . Here we definitively show that the function of CCZ-1 and GLO-3 , but not other gut granule biogenesis factors , is bypassed by the GLO-1 fast-exchange mutant ( Figs 6 and 7 and Tables 1 and 2 ) . Furthermore , GFP::GLO-1 is diffusely localized in ccz-1 ( - ) and glo-3 ( - ) mutants and GLO-3 functions in the recruitment to and/or stabilization of GLO-1 on gut granules ( Fig 8 ) . Loss of GLO-1 GEF activity should result in the accumulation of GLO-1 in the GDP bound form , which would be extracted from organelle membranes into the cytosol by Rab GDI [29 , 31] . Together , these results strongly suggest a CCZ-1 and GLO-3 function as a GEF that activates and localizes GLO-1 . While the GLO-1 ( D132A ) fast-exchange mutant promoted the proper localization of some gut granule proteins in glo-3 ( - ) and ccz-1 ( - ) mutants , the localization of GLO-1 ( D132A ) was reduced and LMP-1 was noticeably absent from these organelles ( Fig 7 ) . Rab GEFs are known to play important roles localizing their Rab substrates , and it is currently unknown whether this is purely through catalyzing nucleotide exchange or through other functions such as physical interactions that recruit the Rab [29 , 31] . Our observations suggest the latter possibility for GLO-3 and CCZ-1 . It is also possible that the higher level of GLO-1 ( D132A ) gut granule association when GLO-3 and CCZ-1 are present , could result from these factors promoting the GTP , membrane localized form of the fast exchange mutant . LMP-1 trafficking to gut granules requires the function of the AP-3 adaptor complex , while other gut granule proteins can localize to gut granules independently of AP-3 [13] . The defects in LMP-1 localization could result from GLO-3 and CCZ-1 functioning in the AP-3 pathway independently of regulating GLO-1 . The Rab GEFs , Rabin8 , VARP , and possibly the TRAPP complexes , have GEF-independent roles in membrane dynamics [93–95] . Additionally , the activation cycle of GLO-1 ( D132A ) might not fully restore wild-type GLO-1 activity , disrupting the delivery of LMP-1 to gut granules . In support of this possibility , biochemical and genetic analysis of analogous fast-exchange mutations in RAB-7 show that they cause decreased RAB-7 function [61 , 96] . Our in vivo , genetic studies suggest that the GEF activity of CCZ-1 for two different Rabs is controlled by its interaction partner . CCZ-1 and SAND-1 ( MON1 ) have amino-terminal longin domains that mediate heterodimerization and nucleotide exchange by RAB-7 [40] . Our in vitro binding studies show that the amino terminal region of CCZ-1 containing the longin domain is sufficient to interact with GLO-3 ( Fig 1 ) . This suggests mutually exclusive binding of GLO-3 and SAND-1 ( MON1 ) to CCZ-1 due to competition for the same binding site on CCZ-1 , which could regulate Rab substrate specificity . GLO-3 has an amino-terminal longin-like domain [19] , however in our assays this region does not show strong interactions with CCZ-1 ( Fig 1 ) . Similar to CCZ-1/SAND-1 ( MON1 ) , the TRAPPII and TRAPPIII complexes are Rab GEFs that are composed of longin domain containing subunits [97 , 98] . Interestingly , it has recently been shown that substitution of longin domain containing subunits can alter the Rab GEF specificity of the TRAPP complexes [99–103] . In the CCZ-1/SAND-1 ( MON1 ) complex , SAND-1 ( MON1 ) makes the majority of contacts with RAB-7 [40] . Thus , substitution of GLO-3 for SAND-1 ( MON1 ) in a heterodimer with CCZ-1 could redirect the specificity of a CCZ-1 containing GEF from RAB-7 to GLO-1 . In support of this idea ccz-1 ( - ) mutants show defects in gut granule protein localization not seen in sand-1 ( - ) , glo-1 ( - ) , or glo-3 ( - ) single mutants ( Fig 4 ) . However , ccz-1 ( - ) is phenocopied by sand-1 ( - ) ; glo-3 ( - ) and sand-1 ( - ) ; glo-1 ( - ) double mutants ( Fig 4 ) . The similar effects of glo-1 ( - ) , glo-3 ( - ) , and ccz-1 ( - ) mutants on gut granule protein localization ( Fig 2 ) [19] , the cytoplasmic mislocalization of GLO-1 in glo-3 ( - ) and ccz-1 ( - ) mutants ( Fig 8 ) , the ability of the fast exchange GLO-1 ( D132A ) mutants to restore gut granules in glo-3 ( - ) and ccz-1 ( - ) mutants ( Figs 6 and 7 and Tables 1 & 2 ) , and the ability of GLO-3 to directly bind CCZ-1 ( Fig 1 ) , strongly support the model that CCZ-1 , by functioning with GLO-3 , can regulate GLO-1 . In mammalian cells the Ccz1/Mon1 ( SAND-1 ) complex acts as a GEF for Rab7 and the HPS1/HPS4 ( BLOC-3 ) complex acts as a GEF for the GLO-1 homologues Rab32 and Rab38 [33 , 43] . C . elegans does not appear to similarly segregate Rab7 and Rab32/38 GEF activities . However , while mammalian Rab32/38 expression is restricted to a subset of cell types [1 , 104–107] , BLOC-3 subunits are ubiquitously expressed [34–36] , and BLOC-3 mutants disrupt normal endolysosome distribution [108] . This supports the possibility that mammalian HPS1 and HPS4 regulate Rabs promoting both LRO and conventional endolysosome biogenesis , similarly to CCZ-1 in C . elegans . glo-3 ( - ) weak mutants show SAND-1 dependent mislocalization of RAB-7 to gut granules ( Figs 10 and S4 and S5 ) . If GLO-3 and SAND-1 compete for CCZ-1 binding then the mislocalization of RAB-7 could result from excessive CCZ-1/SAND-1 ( MON1 ) heterodimer formation and RAB-7 activation when GLO-3 levels are reduced . Alternatively , GLO-3 might inhibit RAB-7 association with gut granules by recruiting or activating a negative regulator of RAB-7 activity . Ectopically expressed GLO-1 reduced the association of RAB-7 with gut granules in glo-3 ( - ) mutants ( Fig 10 ) , suggesting the possibility of a Rab cascade in gut granule biogenesis [30 , 31] . We found that rab-7 ( - ) mutants , despite generating gut granules that phenotypically resemble gut granules in weak glo-3 ( - ) mutants ( Fig 9 ) [19] , properly localize GLO-1 to gut granules ( Fig 10 ) . Disrupting the activity of tbc-2 , which encodes a possible RAB-7 GAP [69–71] , did not lead to the gut granule association of RAB-7 . However , it is possible that RAB-7 acts in concert with other factors to recruit the GLO-1 GEF and that activated GLO-1 recruits a different RAB-7 GAP to mediate an exchange of RAB-7 on late endosomes for GLO-1 on gut granules . Rab GEFs are currently thought to be the major factors determining the subcellular localization of Rabs [33 , 44 , 109–112] . GLO-3 associates with gut granules [44] , putting it in the correct position to direct GLO-1 localization . However , if GLO-3 and CCZ-1 function as a GLO-1 GEF , how does GLO-1 ( D132A ) localize to gut granules in the absence of these proteins ? Analogous fast exchange Rab7 and RAB-2 mutants are properly localized when the activity of their respective GEFs is lacking [113 , 114] , suggesting that GEFs are not essential for the localization of Rabs that have an increased rate of nucleotide exchange . We know little about the identity and function of factors other than GEFs that impact the recruitment and/or stabilization of most Rabs , but they have been suggested to include Rab-GDI displacement factors or Rab effectors [115–118] . It is likely that each Rab utilizes a distinct set of interacting factors and mechanisms to ensure its correct spatiotemporal distribution [29 , 31] . Rab GEFs are typically not membrane anchored , a key characteristic of a membrane targeting receptor . Possibly there are integral membrane proteins that function as Rab receptors or modify the organelle membrane to promote Rab localization . The identification and characterization of these factors will be critical for understanding how organelles acquire their functional identity .
C . elegans strains were cultured at 22°C on NGM media seeded with E . coli strain OP50 [119] . N2 was used as the wild type and all mutant alleles were generated in this strain . The following mutations were used: apt-7 ( tm920 ) , ccz-1 ( ok2182 ) , cup-5 ( zu223 ) , glo-1 ( zu437 ) , glo-2 ( tm592 ) , glo-3 ( gk582755 ) , glo-3 ( kx38 ) , glo-3 ( kx94 ) , glo-3 ( syb272 ) , glo-3 ( zu446 ) , glo-4 ( ok623 ) , rab-7 ( ok511 ) , sand-1 ( or552 ) , snpn-1 ( tm1892 ) , tbc-2 ( tm2241 ) , unc-36 ( e251 ) , vps-18 ( tm1125 ) . Wormbase ( www . wormbase . org ) hosts descriptions of each allele . The following transgenes were used: amIs4[cdf-2p::cdf-2::gfp; unc-119 ( + ) ] [46] , cbgIs98[pept-1p::gfp::rab-11 . 1; unc-119 ( + ) ] [120] , kxEx9[glo-1p::gfp::glo-1; Rol-6D] [1] , kxEx141[cpr-6p::cpr-6::mCherry; Rol-6D] [13] , kxEx148[gba-3p::gba-3::mCherry; Rol-6D] [13] , kxEx223[glo-1p::gfp::glo-1 ( T25N ) ; Rol-6D] ( this work ) , kxEx230[glo-1p::gfp::glo-1 ( Q71L ) ; Rol-6D] ( this work ) , kxEx252[vha-6p::gfp::glo-1 ( D132A ) ; Rol-6D] [19] , kxEx254[vha-6p::gfp::glo-1 ( I133F ) ; Rol-6D] [19] , kxEx272[glo-1p::gfp::glo-1 ( D132A ) ; Rol-6D] ( this work ) , kxEx273[glo-1p::gfp::glo-1 ( I133F ) ; Rol-6D] ( this work ) , pwIs50[lmp-1p::lmp-1::gfp; unc-119 ( + ) ] [55] , pwIs72[vha-6p::gfp::rab-5; unc-119 ( + ) ] [1] , pwIs170[vha-6p::gfp::rab-7; unc-119 ( + ) ] [121] , pwIs206[vha-6p::gfp::rab-10; unc-119 ( + ) ] [121] , pwIs503[vha-6p::mans::gfp; unc-119 ( + ) ] [121] , tdEx2[arl-8p::arl-8::gfp; Rol-6D] [52] . Integrated ( Is ) and extrachromosomal ( Ex ) transgenes , present in otherwise wild-type strains , were moved into mutant backgrounds by crossing hermaphrodites containing the transgenes with males homozygous or heterozygous for the mutation . The presence of the mutation in the resulting strain was confirmed by the presence of the mutant phenotype , or in cases where this was modified by the transgene , by PCR and/or DNA sequencing . To generate double mutants containing glo-3 ( kx38 ) , transheterozygous individuals were allowed to self fertilize and progeny that were homozygous for glo-3 ( kx38 ) , as evidenced by the number of birefringent gut granules , and heterozygous for the other mutation , were isolated . The homozygous double mutants that exhibited the other mutant phenotype were then isolated from these strains . In cases where one Glo phenotype masked another , we confirmed the presence of the masked mutation using PCR/DNA sequencing . In all cases , single and double mutants were homozygous for each mutation except strains containing rab-7 ( ok511 ) , cup-5 ( zu223 ) , and some strains containing ccz-1 ( ok2182 ) , which were kept heterozygous due to the recessive maternal effect lethality or severe growth defect caused by these mutations [54 , 57 , 86] . In cases where strains heterozygous for these mutations were used , we identified homozygous mutant adults by the presence of large DIC refractile granules within embryos in their uterus or a linked recessive marker . Mutant embryos produced by homozygous rab-7 ( - ) , cup-5 ( - ) , and ccz-1 ( - ) parents display these morphologically distinct structures [54 , 57 , 86] . unc-36 ( e251 ) was linked to cup-5 ( - ) and we found that it did not alter gut granule biogenesis in any of our assays . glo-3 ( gk582755 ) was identified in an ongoing screen of Million Mutation strains for defects in gut granule number and/or morphology . The Glo phenotype of strain VC40338 mapped to the X chromosome and did not complement the Glo phenotypes of glo-3 ( zu446 ) . The glo-3 ( gk582755 ) mutation causes a GLO-3 ( N279K ) substitution and was backcrossed 3 times to N2 before being characterized . CRISPR-Cas9 gene editing was carried out by SunyBiotech ( Fuzhough City , Fujian , China ) to generate glo-3 ( syb272 ) , which precisely removes the entire glo-3 coding sequence . Sanger sequencing verifying the presence of the deletion was carried out by Genewiz ( South Plainfield , NJ , USA ) . The resulting sequence TTCgAGGTAAACTCGTTCAAA—ATAATTTATATTTACAAGTAT flanked the deletion ( marked by— ) . The g denotes a mutation created to destroy the PAM site . syb272 was backcrossed 4 times to N2 before being characterized . To knock down the expression of rab-7 we used RNAi feeding protocol 1 described in [122] and clones from the Ahringer RNAi library ( Source Bioscience , Nottingham , UK ) . The effects of rab-7 RNAi were not seen in embryos treated with F33E2 . 4 ( RNAi ) , which targets a gene not required for gut granule biogenesis . In RNAi experiments , inhibition of rab-7 activity was confirmed by the presence of DIC refractile granules . Widefield polarization and fluorescence microscopy was carried out with a Zeiss AxioImager . M2 and images were captured with an AxioCam MRm digital camera controlled by AxioVision 4 . 8 software ( Zeiss , Thornwood , NY ) . Confocal fluorescence microscopy was carried out with a Zeiss LSM710 laser scanning confocal microscope . Embryos were imaged with 100X or 63X Plan-Apochromat 1 . 4 NA objectives and adults were imaged with a 40X Plan-Apochromat 1 . 3 NA objective . Adults were mounted on 3% agarose pads and immobilized with 10mM levamisole ( Sigma Aldrich , St . Louis , MO ) . Autofluorescent gut granules were imaged with a Zeiss 38 filter ( GFP , excitation , BP 470/40; emission , BP 525/50 ) , a Zeiss 45 filter ( mCherry , excitation , BP 560/40; emission , BP 630/75 ) , or a 488 laser line . Z-stacks of the intestine were captured and maximum intensity projections of ½ or the entire depth of the intestine are shown . In adults expressing GFP tagged proteins , the Zeiss 45 filter was used to visualize gut granules . Living embryos were mounted in H2O on 3% agarose pads . Body movements in embryos do not begin until after the 1 . 5-fold stage . To acquire images of late stage embryos , excess respiring OP50 bacteria was added to induce hypoxia and immobilization . Birefringent material was visualized with polarization optics . Maximum intensity projections of Z-stacks capturing all of the birefringent material within the intestine are shown . GFP and mCherry markers were imaged by confocal microscopy in living 1 . 5-fold stage embryos using the 488 and 561 laser lines . GFP , mCherry , and autofluorescence in living embryos were imaged using widefield microscopy with Zeiss , 38 , Zeiss 45 and Zeiss 49 ( DAPI , excitation , G 365; emission , 445/50 ) fluorescence filters , respectively . To characterize the pattern of GFP::GLO-1 in living embryos , GFP::GLO-1 signals in each strain were captured using widefield fluorescence microscopy using identical exposure settings . Z-stacks through the top half of the intestine were captured . The Fiji software plot profile tool centered on randomly selected puncta was used to generate intensity profile histograms [123] . The intensity value for each punctum was calculated by subtracting the average of the 10 lowest intensity values from the peak value in each 60 pixel intensity histogram . Widefield fluorescence Z-stacks of autofluorescent gut granules in pretzel stage embryos were imaged with a Zeiss 49 filter . The diameter of these organelles was determined using Zeiss AxioVision software . Embryos were fixed in -20°C MeOH following a freeze-crack as described [124] . The intrinsic fluorescence of GFP was used to visualize the distribution of GFP tagged proteins after fixation . Antibodies to GLO-1 [1] , LMP-1 [125] , PGP-2 [47] , RAB-5 [126] , and RAB-7 [127] were used . Z-stacks through the intestine of fixed LMP-1::GFP expressing embryos were imaged using widefield microcopy with a Zeiss 38 filter . Using the plasma membrane localization of LMP-1::GFP to identify individual cells , we manually quantified the number of lysosomes located within the 4 cells that make up Int 2 and 3 . To simultaneously image the localization of CDF-2::GFP , PGP-2 , and LMP-1 in embryos , we used secondary antibodies marked with DyLight 405 and Rhodamine Red ( Jackson ImmunoResearch , West Grove , PA ) . These were imaged with confocal microscopy using the 405 , 488 , and 561 laser lines . Widefield fluorescence microscopy was used in some experiments to capture GFP , Alexa 488 , or Rhodamine Red fluorescence with Zeiss 45 or Zeiss 49 filters . The number of PGP-2 marked organelles in individual embryos was quantified using SQUASSH software analysis of confocal Z-stacks spanning the entire intestine [64] . Using confocal Z-stacks , the diameter of CDF-2::GFP or anti-LMP-1 marked organelles was determined using Zeiss Zen Blue 2012 software . In glo-3 ( kx38 ) containing strains , only the diameter of CDF-2::GFP organelles that also contained PGP-2 were measured , as CDF-2::GFP was mislocalized to non-PGP-2 containing lysosomes in glo-3 ( kx38 ) mutants . To determine endolysosome size , only LMP-1 containing organelles that lacked the gut granule marker PGP-2 were measured . For colocalization studies , two or three channel confocal Z-stacks were acquired and analyzed . As noted in the figure legends , either manual or automated colocalization scoring was performed . In some experiments , randomly selected organelles labeled by one marker were manually scored for the presence of a second marker . Individual organelle signals were scored as colocalizing if they overlapped by more than 50% . The colocalization per embryo was calculated and these values were used to determine the mean colocalization shown in the graphs . In other experiments , SQUASSH software was used to segment the Z-stack and every identified organelle was used in the analysis ( typically 120–160 gut granules per wild-type embryo ) . The resulting Csize measurement of colocalization represents the fraction of the total volume of one marker that overlapped with the second marker [64] . For example , Csize ( PGP-2LMP-1/PGP-2 ) refers to the area of PGP-2 that overlapped with LMP-1 divided by the total area of PGP-2 , and represents the proportion of PGP-2 that colocalizes with LMP-1 . The Csize per embryo was calculated and these values were used to determine the mean colocalization shown in the graphs . One way ANOVAs were carried out in Microsoft Excel for Mac 2011 . Bonferroni or Tukey-Kramer post hoc tests were used when making 3 or more comparisons . Bar graphs were generated with Excel for Mac 2011 and dot plots were made with R ( version 3 . 1 . 2 ) Beeswarm ( Version 0 . 1 . 6 ) . Figures were constructed with Photoshop CS2 and representative images used to determine marker colocalization and organelle presence , number , or size is shown . Brightness and contrast adjustments were uniformly applied to each panel . The S . cerevisiae EGY48 strain was used for all 2-hydrid assays [128] . The DupLEX-A yeast 2-hybrid system was used according to the manufacturer’s instructions ( Origene Technologies , Rockville , MD , USA ) . The bait plasmids pEG202 and pEG202-NLS encoding LexA DNA binding domain fusions and prey plasmid pJG4-5 encoding B42 transcription activation domain fusions were used . A full-length glo-3 cDNA was PCR amplified from pDONR/Zeo-glo-3 using ( italics are homologous to vector sequences and bold hybridize with the coding sequence ) P1129 5’CAGATTATGCCTCTCCCGCCATGTTTGGTTATGTTGTTGTTAATGAAC3’ and P1130 5’GCGAAGAAGTCCAAAGCTTCGGTTATTTTAACTGTTTTAACACGCATTCC3’ with Q5 High Fidelity DNA polymerase ( NEB , Ipswich , MA , USA ) and inserted into pJG4-5 digested with EcoRI and XhoI using NEBuilder HiFi DNA Assembly Cloning as described by the manufacturer ( NEB ) . A full-length ccz-1 cDNA was amplified from pDONR/Zeo-ccz-1 using P1119 ‘AACGGCGACTGGCTGGCCATGGAGTCGATTGCAAATCCATTG3’ and P1120 5’CTTGGCTGCAGGTCGACGGTCAACTAAAAAATATGGCTTCGAAATGGG3’ with Q5 High Fidelity DNA polymerase ( NEB ) and inserted into pEG202 digested with EcoRI and XhoI , using NEBuilder HiFi DNA Assembly Cloning as described by the manufacturer ( NEB ) . Sequencing of the resulting plasmids showed that the coding sequences lacked mutations and were in-frame with the DNA binding or transcription activation domains ( Genewiz , South Plainfield , NJ , USA ) . Lithium acetate mediated transformation was used to simultaneously introduce combinations of plasmids into EGY48 . LEU2 reporter expression was assessed by growing strains in 2% dextrose lacking histidine , tryptophan , and uracil liquid media overnight , diluting to 1 . 0 OD600 , and spotting this and serial dilutions on 2% dextrose or 2% galactose/1% raffinose plates lacking leucine , histidine , tryptophan , and uracil at 30°C for 3 days . The pSH18-34 reporter plasmid was used to assess lacZ expression by growing strains on 2% dextrose or 2% galactose/1% raffinose plates containing 80μg/ml X-Gal and lacking histidine , tryptophan , and uracil at 30°C for 3 days . Full length ( 1–1821 bp ) and amino terminal encoding ( 1–657 bp ) glo-3 cDNAs were inserted into pGEX4T1 with BamHI and XhoI . Full length ( 1–1584 bp ) and amino terminal encoding ( 1–600 bp ) ccz-1 cDNAs were inserted into pET28a with BamHI and XhoI . Recombinant GST-GLO-3 proteins were expressed in Rosetta ( DE3 ) bacterial cells and purified with glutathione-Sepharose beads ( GE Healthcare Bio-Sciences Pittsburg , PA ) according to the instructions provided by the supplier . Recombinant His6-CCZ-1 proteins were expressed in Rosetta ( DE3 ) bacterial cells and purified with Chelating Sepharose Fast Flow ( GE Healthcare Bio-Sciences Pittsburg , PA ) according to the instructions provided by the supplier . Purified GST or GST-GLO-3 proteins ( 2 . 5 μg of each ) were immobilized on glutathione-Sepharose beads and then pre-incubated with blocking buffer ( 5% BSA , 100mM Tris-HCl PH7 . 5 , 150mM NaCl , 10mM DTT , 0 . 05% NP40 , 1mM PMSF ) at 4°C for 1h , then incubated with His6-CCZ-1 proteins in the binding buffer ( 1%BSA , 100mM Tris-HCl PH7 . 5 , 150mM NaCl , 10mM DTT , 0 . 05% NP40 , 1mM PMSF ) at 4°C for 4 h . After extensively washing with washing buffer ( 100mM Tris-HCl PH7 . 5 , 150mM NaCl , 10mM DTT , 0 . 05% NP40 , 1mM PMSF ) , bound proteins were resolved on sodium dodecyl sulphate ( SDS ) polyacrylamide gels ( SDS-PAGE ) and visualized by Western Blot . Mutations predicted to activate ( Q71L ) or inactivate ( T25N ) GLO-1 were generated using site-directed mutagenesis with Quickchange II ( Agilent Technologies , Santa Clara , CA ) . The vha-6p::gfp::glo-1::let-858 3’UTR plasmid was used as the template for these reactions [1] . The primers P746 5’GGTGATCCAGGTGTCGGTAAAAACTCTATTATTCGTCG3’ and P747 5’CGACGAATAATAGAGTTTTTACCGACACCTGGATCACC3’ were used to generate GLO-1 ( T25N ) and primers P744 ‘CTGGGATATTTCAGGCCTCGACCGATATGGGGTCATG3’ and P745 5’CATGACCCCATATCGGTCGAGGCCTGAAATATCCCAG3’ were used to generate GLO-1 ( Q71L ) . Underlined nucleotides denote the point mutations . The desired mutations , and lack of other mutations in the glo-1 coding sequence , were confirmed by DNA sequencing . The 2 . 1kb glo-1 promoter was added to gfp::glo-1 ( + ) , gfp::glo-1 ( T25N ) , gfp::glo-1 ( Q71L ) , gfp::glo-1 ( D132A ) , and gfp::glo-1 ( I133F ) using PCR fusion [129] . The glo-1 promoter was PCR amplified using genomic DNA as a template with primers P231 5’AACCCAAGCTTCCGTATCTTCTCTCCTTATTTCGACCG3’and P268 5’CAGTGAAAAGTTCTTCTCCTTTACTCATTTTGTTCTGAATATATATTAAAATTAG3’ . gfp::glo-1 ( wild-type or mutant ) ::let-858 3’UTR was PCR amplified from plasmid templates using primers P269 5’ATGAGTAAAGGAGAAGAACTTTTCACTG3’and P500 5’ATTTCCCCGAAAAGTGCCACCTGACG3’ . The promoter was added to the glo-1 coding sequences with primers P265 5’ATAATGGGAACCTGAAATTAGAAGAGG3’ and P271 5’GACTAGTTTTCCTTCCTCCTCTATAT3’ . The resulting fusion products were injected at 1ng/μl with the dominant Rol-6 containing plasmid pRF4 at 100ng/μl . In all cases , multiple independent transgenic lines , which showed the same expression pattern , were isolated for each version of glo-1 . Single arrays were chosen and crossed into different mutant backgrounds . In all studies with embryos , glo-1 expression was controlled by its own promoter . Both the glo-1 and vha-6 promoters express in adult intestinal cells and thus in studies with adults , glo-1 expression was controlled by either promoter . | Lysosome-related organelles represent a diverse collection of intracellular compartments that have important physiological functions . In mammals they include melanosomes , key sites of pigment synthesis , and platelet dense granules , which play important roles in blood clotting . Defects in the pathways that direct the formation of these LROs cause Hermansky-Pudlak syndrome , which is characterized by partial albinism and excessive bleeding . Here we describe studies of an LRO found in Caenorhabditis elegans , whose formation is mediated by evolutionarily conserved machinery and pathways . Our work supports a novel model for how a key player in LRO formation , the Rab32 orthologue GLO-1 , is regulated and functions in C . elegans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"lysosomes",
"caenorhabditis",
"light",
"microscopy",
"animals",
"animal",
"models",
"developmental",
"biology",
"guanine",
"nucleotide",
"exchange",
"factors",
"caenorhabditis",
"elegans",
"model",
"organisms",
"microscopy",
"experimental",
"organism",
"systems",
"confocal",
"microscopy",
"embryos",
"cellular",
"structures",
"and",
"organelles",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"embryology",
"animal",
"studies",
"gastrointestinal",
"tract",
"biochemistry",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"anatomy",
"nematoda",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"cell",
"signaling",
"organisms",
"signaling",
"molecules"
] | 2018 | Function and regulation of the Caenorhabditis elegans Rab32 family member GLO-1 in lysosome-related organelle biogenesis |
A recent paper of B . Naundorf et al . described an intriguing negative correlation between variability of the onset potential at which an action potential occurs ( the onset span ) and the rapidity of action potential initiation ( the onset rapidity ) . This correlation was demonstrated in numerical simulations of the Hodgkin-Huxley model . Due to this antagonism , it is argued that Hodgkin-Huxley-type models are unable to explain action potential initiation observed in cortical neurons in vivo or in vitro . Here we apply a method from theoretical physics to derive an analytical characterization of this problem . We analytically compute the probability distribution of onset potentials and analytically derive the inverse relationship between onset span and onset rapidity . We find that the relationship between onset span and onset rapidity depends on the level of synaptic background activity . Hence we are able to elucidate the regions of parameter space for which the Hodgkin-Huxley model is able to accurately describe the behavior of this system .
In 1952 , Hodgkin and Huxley explained how action potentials are generated through the electrical excitability of neuronal membranes [1] . Action potentials arise from the synergistic action of sodium channels and potassium channels , each of which opens and closes in a voltage dependent fashion . A key feature of their model is that the channels open independently of each other; the probability that a channel is open depends only on the membrane voltage history . A recent paper [2] challenged this picture . Therein the dynamics of action potential initiation in cortical neurons in vivo and in vitro are analyzed . The authors focus on two variables , the onset potential , i . e . the membrane potential at which an action potential fires , and the onset rapidity , or rate with which the action potential initially fires . Naundorf et al . argue that the variability or span of onset potentials observed in experiments , in conjunction with their swift onset rapidity , cannot be explained by the Hodgkin-Huxley model . In particular , within the Hodgkin-Huxley model they demonstrate through numerical simulations an antagonistic relationship between these two variables . If parameters are adjusted to fit the onset rapidity of the data , the observed onset span disagrees with the model , and vice versa . To fix this discrepancy [2] argues for a radical rethinking of the basic underpinnings of the Hodgkin and Huxley model , in which the probability of an ion channel being open depends not only on the membrane potential but also on the local density of channels . The result reported in [2] was critically analyzed in a recent letter of D . A . McCormick et al . [3] . In [3] it was proposed that the observed combination of large onset span and swift onset rapidity could be captured using a Hodgkin-Huxley model if action potentials were initiated at one place within the cell , ( the axon initial segment ) , and then propagated around 30 microns to the site at which they were recorded , ( the soma ) . Whole-cell recordings from the soma of cortical pyramidal cells in vitro demonstrated faster onset rapidity and larger onset span then those obtained from the axon initial segment . This seemingly compelling reappraisal of the original data was in turn dissected by Naundorf et al . in [4] where it is suggested that the physiological setting of [3] is unrealistic , and the model inadequate . Here we use a standard technique from theoretical physics ( the path integral ) to derive an analytical formula relating the onset rapidity and onset span . Our analysis applies to the classical Hodgkin-Huxley model , in addition to generalizations thereof , including those in which the channel opening probability depends on channel density [2] . To derive an analytical characterization of this relationship , we directly compute the probability distribution of the onset potential and demonstrate how it depends on model parameters . The formula that we arrive at can be used to compare experimental observations with the parameter values incorporated into such models . As anticipated by [2] , a broad class of ion channel models displays an inverse relationship between onset rapidity and onset span . We find that the parameter relating onset rapidity to onset span depends on the amount of synaptic background activity included in the model . Indeed , a range of background activity exists where the classical Hodgkin-Huxley model agrees with the experimental data reported in [2] .
We first review the essential framework of Hodgkin-Huxley type models for action potential generation . The dynamics of the membrane potential of a section of neuron , assumed to be spatially homogeneous , are given by [1]: ( 1 ) whereHere is the membrane capacitance , is the maximal conductance of channels of type , is the probability that a channel of type is open , is the reversal potential for channel type and the subscripts , and refer to sodium , potassium and M-type potassium channels respectively . A leak current is included with conductance and reversal potential , is the membrane area , while is the current resulting from synaptic background activity [5] . Background activity is typically modeled by assuming synaptic conductances are stochastic and consists of an excitatory conductance with reversal potential and an inhibitory conductance with reversal potential , as found in [6] so that ( 2 ) In [2] the conductances and are modeled by Ornstein-Uhlenbeck processes with correlation times and , and noise diffusion coefficients and respectively [7] . We are interested in understanding from this model the relationship between onset span and onset rapidity , as defined by [2] . As described above , the onset rapidity is the rate at which the voltage increases; near onset the increase in voltage is exponential and so is given by the slope of a plot of versus . The onset span measures the variability of the voltage threshold for action potential initiation , [2] defines this threshold as the voltage at which , and takes . Due to the stochastic synaptic background , there is a distribution of voltages at which the voltage threshold is attained; the onset span is given by the width of this distribution . We calculate the probability distribution of voltage thresholds , and derive the onset span from the moments of this distribution .
To proceed we use the fact that , at action potential initiation , we need only consider the sodium channels . This is because the potassium channels respond too slowly for their dynamics to influence the voltage [8] . Moreover , near threshold , the probability that a sodium channel is open depends only on the membrane voltage . This probability is traditionally measured by the so-called activation curve [9] , where . Under these assumptions , Eq . ( 1 ) reduces to ( 3 ) Action potential onset occurs when reaches , where is an unstable equilibrium of Eq . ( 3 ) in the absence of noise . Below the membrane potential relaxes to its resting potential , whereas above an action potential fires . To study the dynamics near onset , we therefore write , and expand equation ( 3 ) to leading order in , obtaining ( 4 ) whereandWe use the parameter values and as found in [6] and used in [2] . Thus . Near threshold the synaptic background itself is a single gaussian noise source with diffusion constant characterized byNote that in equation ( 4 ) , is the onset rapidity . According to [2] , the voltage threshold is defined as the voltage at which , where denotes the time derivative of . Owing to the noise source there is a range of values at which this condition is attained . The onset span describes the range observed , and is related to the standard deviation of the probability distribution for these voltage thresholds . Consider trajectories subject to the boundary conditions and , where is the time at which the voltage threshold is attained . There is a distribution of times at which the threshold condition can be met . Moreover , for a given , there is a distribution of voltages that the trajectory might attain at time . This distribution is characterized by a mean , as well as a variance . The total variance of the voltage threshold is therefore given by ( 5 ) where is the probability that the voltage threshold occurs at time , and denotes the expectation . The first two terms of equation ( 5 ) make up the variance of mean values that occur owing to the range of times at which the threshold condition is met . For each such time , the final term sums the variance of voltages likely to be reached about the mean value . Equation ( 5 ) is the fundamental equation for the onset span: it requires us to compute , and . To proceed , we use the fact that the noise source is Gaussian with variance , and therefore the probability density of a given realization of the noise between isThis leads to a path integral formulation of the probability of realizing a particular trajectory with , as developed in [10] . As equation ( 4 ) implies , we find ( 6 ) Here the integral is taken over all the possible paths that might take between time and . Some paths are of course more likely then others; application of the Euler-Lagrange equation finds that the most probable trajectory of Eq . ( 6 ) is the saddle point . It minimizessubject to the boundary conditions and and therefore satisfiesThe most probable trajectory is the minimum of this quantity by definition . Since the probability density is of the form , where is positive definite , the trajectory that minimizes maximizes the probability . Imposing the boundary conditions we have ( 7 ) We can use insert this solution into Eq . ( 6 ) , in order to compute the probability density of this trajectory occurring . We obtain ( 8 ) It is convenient to rewrite this formula by defining the dimensionless parameters and . Since is a monotonic function of and thus also of we can transform this to the probability density that the voltage threshold is achieved at time , namely ( 9 ) In Eqs . ( 8 ) and ( 9 ) the constants and are set by the normalization condition . We have now computed two of the three quantities needed to evaluate Eq . ( 5 ) for the onset span . Thus we are able to evaluate the first two terms of this equations . Our theory has captured the probability distribution of the mean , but we also need to compute the variance about this mean in order to fully evaluate Eq . ( 5 ) for . We can calculate this variance by noting that a general solution that satisfies and can be written as , where can be expanded in the Fourier series Substituting this into Eq . ( 6 ) , we obtain ( 10 ) where is a normalization constant . This demonstrates that the total probability distribution is a product of the probability for the mean trajectory , with Gaussian probability distributions for each of the . Now , Eq . ( 10 ) shows that each has mean zero and varianceHence the variance of is given by ( 11 ) where we have again used the dimensionless parameters and as defined above . We now can evaluate Eq . 5 for . Taking Eqs . ( 7 ) , ( 9 ) and ( 11 ) and letting we have ( 12 ) ( 13 ) The first two terms of Eq . ( 12 ) are the variance of the voltages reached by the mean path , for each time at which the threshold might be reached . The last term adds in the variance about the mean path for each value of , that is the variability from . Equation ( 12 ) is the central result of this paper , directly relating the onset span to the noise strength , the voltage threshold and the onset rapidity . Figure 1 shows a numerical evaluation of . Asymptotic analysis of the integral in Eq . ( 12 ) shows that at small , , and at large , ( Figure 1 ) . Hence we obtain ( 14 ) ( 15 ) We note that the low limit describes the behavior of a simple random walk; here a small value of corresponds to a low threshold for the derivative . Thus the variance of onset voltages is simply the variance of all possible trajectories the random walk might take . In the high limit the size of the noise term ceases to much affect the variance of onset voltages . As the derivative threshold is high in this case , the deterministic exponential growth behavior will dominate those trajectories that reach the threshold . Thus we have calculated the variance of voltages at which action potential onset occurs as a function of the onset rapidity , the onset threshold and the level of synaptic background activity present . In Figure 2A we have simulated a pair of trajectories with parameter values and , and in Figure 2B a pair with parameter values and . To ascertain the onset potential of each simulated trajectory we need to find the voltage at which the derivative of the trajectory first exceeds the threshold . As the model in ( 4 ) is not differentiable , it is necessary to fit a ‘smoothed’ curve to each trajectory , and find the derivative of this curve . In Figure 2C and 2D we have fitted an exponential curve with equation to each simulated trajectory . Figure 2E and 2F show the derivative extracted as a function of the voltage . To demonstrate the validity of our analysis , we use the reduced Hodgkin Huxley model described by ( 4 ) to simulate trajectories and compare the onset span we observe for particular sets of parameter values with that predicted by our analysis . In order to simulate the gaussian noise source in Eq . ( 4 ) we use a Wiener process with the appropriate diffusion constant . In Figure 3 we choose two sets of parameter values and compare the range of onset potentials found by simulation with that predicted by our analysis . The black stars are the points at which each trajectory crossed the derivative threshold . On each plot the endpoints were grouped into bins of width . The average voltage in each bin is plotted in magenta , while the mean onset voltage at the center of each bin as predicted by our analysis is plotted in red . Similarly the standard deviation about the mean in each bin is plotted in cyan , and can be compared with the standard deviation predicted by our analysis which has been plotted in green . We observe that both the mean onset potential and the standard deviation about the mean at each time point found in the simulations is well matched by that predicted by our analysis . In both the low limit and the high limit we found in Eq . ( 14 ) that there is indeed an antagonistic relationship between and , as argued by Naundorf et al . [2] . They observed that changing the parameters of the activation curve and the peak sodium conductance led to antagonistic changes in the onset rapidity and the onset span; hence they were not able to fit the Hodgkin-Huxley model to their data . Equations ( 14 ) and ( 15 ) show that the antagonistic relationship between and is controlled by in the limit of low , and in the limit of high . Neither ( the variance of the synaptic noise strength ) nor ( the criterion for the voltage threshold ) were varied in the simulations of Naundorf et al . [2] . We observe that our analysis can also be applied to the cooperative model proposed in [2] , in which the probability of channel opening depends on both the membrane voltage , and the local channel density . In the vicinity of the unstable fixed point , incorporating the local channel density alters the value of , but does not change the form of equation ( 4 ) . We now compare the theory to the results of Naundorf . In their experiments , they measure the onset span as the difference between the maximum and minimum voltage threshold that is measured . Since 99 . 7% of observations fall within three standard deviations of the mean , we can approximate the onset span of between 50 and 500 trials as six times the standard deviation . We assume that the calculation of the onset span from the simulations in Naundorf was done in the same fashion . In Figure 4 we have calculated the onset span as a function of using different values of . Changing the noise strength allows the theoretical curves to move between the various regimes observed experimentally . For most of the curves through the experimental data , a noise diffusion constant of around to fits the data well . Although this is a larger diffusion constant than that apparently used in the simulations of Naundorf et al . , this value does a good job of emulating the experimental trajectories shown in Figure 2C and 2D of [2] . Figure 2B shows a simulated trajectory with noise strength while Figure 2A shows a simulation with a smaller diffusion coefficient of . The voltage trace at is visually similar to the behavior in Figure 2B of Naundorf in the vicinity of the unstable fixed point , whereas Figure 2A does not compare well , the noise level is much too low . Note that because we have linearized around the unstable fixed point , we can only expect to capture the behavior around the voltage threshold; this is presumably the reason that our simulations in Figure 2 do not reproduce the vertical spiking behavior occurring after action potential onset in Figure 2B of [2] . It is worth noting that additional sources of variance exist when comparing the experiments to the theory . In particular , ( i ) the theory assumes that the voltage threshold occurs precisely when in contrast the experimental data show substantial variability in . Additionally ( ii ) experiments report an averaged onset rapidity , whereas our analysis indicates a direct relationship between the onset potential and . Both factors ( i ) and ( ii ) artificially increase the onset span . The calculations described here clarify that to understand whether the experimental data is consistent with the Hodgkin Huxley picture , it is necessary to understand the corresponding level of ; ideally , independent measurements of the synaptic background statistics are required . Intense levels of background activity characterized by high amplitude membrane potential fluctuations are known to occur during active states in neocortical neurons [11] . Combining the theoretical formalism described herein with measurements of the variance of synaptic conductances [12] , while carefully controlling for other sources of variability in the measurement , is an excellent direction for future research . | In 1952 , Hodgkin and Huxley described the underlying mechanism for the firing of action potentials through which information is propagated in the nervous system . Hodgkin and Huxley's model relies on the opening and closing of channels , selectively allowing ions to move across the membrane . In the original picture , the channels open independently of one another . A recent paper argues that this model is incapable of modeling a set of action potential data recorded in the cortical neurons of cats . Instead the authors suggest that to model their data it is necessary to conclude that ion channels open cooperatively , so that opening one channel increases the chance that another channel opens . We analyze the initiation of action potentials using a method from theoretical physics , the path integral . We demonstrate that deviations of the data from the predictions of the Hodgkin-Huxley model hinge on measurement of the noise strength . | [
"Abstract",
"Introduction",
"Model",
"Results/Discussion"
] | [
"mathematics",
"biophysics/theory",
"and",
"simulation",
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2009 | Action Potential Initiation in the Hodgkin-Huxley Model |
Mycobacterium ulcerans , the causative agent of Buruli ulcer ( BU ) , is unique among human pathogens in its capacity to produce a polyketide-derived macrolide called mycolactone , making this molecule an attractive candidate target for diagnosis and disease monitoring . Whether mycolactone diffuses from ulcerated lesions in clinically accessible samples and is modulated by antibiotic therapy remained to be established . Peripheral blood and ulcer exudates were sampled from patients at various stages of antibiotic therapy in Ghana and Ivory Coast . Total lipids were extracted from serum , white cell pellets and ulcer exudates with organic solvents . The presence of mycolactone in these extracts was then analyzed by a recently published , field-friendly method using thin layer chromatography and fluorescence detection . This approach did not allow us to detect mycolactone accurately , because of a high background due to co-extracted human lipids . We thus used a previously established approach based on high performance liquid chromatography coupled to mass spectrometry . By this means , we could identify structurally intact mycolactone in ulcer exudates and serum of patients , and evaluate the impact of antibiotic treatment on the concentration of mycolactone . Our study provides the proof of concept that assays based on mycolactone detection in serum and ulcer exudates can form the basis of BU diagnostic tests . However , the identification of mycolactone required a technology that is not compatible with field conditions and point-of-care assays for mycolactone detection remain to be worked out . Notably , we found mycolactone in ulcer exudates harvested at the end of antibiotic therapy , suggesting that the toxin is eliminated by BU patients at a slow rate . Our results also indicated that mycolactone titres in the serum may reflect a positive response to antibiotics , a possibility that it will be interesting to examine further through longitudinal studies .
Buruli ulcer ( BU ) , caused by Mycobacterium ulcerans , is the third most common mycobacterial disease after tuberculosis and leprosy and represents an emerging threat [1] , [2] . Since the late 1980s , the disease has been developing throughout West and Central Africa , prompting the WHO in 1998 to initiate an awareness and control campaign ( http://www . who . int/gtb-buruli ) . Although efficient [3] , [4] , [5] , [6] , [7] , current treatment protocols recommend the daily administration of oral rifampicin and intramuscular streptomycin for 8 weeks , with additional surgical intervention when necessary . To control the emergence of BU and to improve the management of the disease , it is vital to develop new tools for early diagnosis and treatment monitoring . A distinctive feature of M . ulcerans among human pathogens is the production of mycolactone [8] , a macrocyclic polyketide playing a critical role in bacterial virulence ( reviewed in [9] , [10] ) . We have recently demonstrated the presence of intact mycolactone in punch biopsies from all forms of BU disease , before and during antibiotic therapy [11] . Moreover , there is evidence from mouse studies that mycolactone may diffuse into the peripheral blood [12] . Here we used chemical approaches to determine if mycolactone is present in blood samples and ulcer exudates obtained non-invasively at various stages of antibiotic therapy .
Patients were recruited if they met the WHO clinical case definition of BU disease; were not pregnant; had no history of tuberculosis , leprosy , or liver , kidney , or hearing impairment . All subjects provided written informed consent ( thumb print of parent or guardian in the case of children , depending on literacy ) . A cross-section of patients with BU disease were recruited which included a spectrum of patients yet to initiate antibiotic therapy , some at various stages of antibiotic treatment and few who had completed treatment . Healthy controls from the same endemic area were included . In Ghana , patients were recruited by local health workers from villages near Tepa Government Hospital in the Ahafo Ano North District of Ghana , where there is a high prevalence of BU . The study protocol was approved by the ethics review committees at the School of Medical Sciences , Kwame Nkrumah University of Science and Technology , Kumasi , Ghana . In Ivory Coast , patients were either recruited from the Djekanou General Hospital or detected by a mobile medical team actively screening the district of Abidjan . The study protocol was approved by the national ethic review committee . To confirm the clinical diagnosis , punch biopsy specimens of 4-mm diameter ( Ghana ) or ulcer exudates ( Ivory Coast ) were tested by PCR for the IS2404 repeat sequence , which is characteristic of M . ulcerans . Positive patients were treated with 10 mg/kg oral rifampicin and 15 mg/kg intramuscular streptomycin daily , administered at village health posts under direct observation , according to the WHO recommendations . Only IS2404 positive samples were considered for analysis of mycolactone presence . Mycolactone was extracted from M . ulcerans 1615 ( ATCC 35840 ) , as previously described [12] . In brief , bacteria were cultivated in Middlebrook 7H9 broth ( Difco ) enriched with 10% oleic acid-albumin-dextrose-catalase ( OADC , Becton Dickinson ) for 4 weeks in spinner flasks at 30°C . Total lipids were extracted from bacterial cell pellets with 2/1 CHCl3/MeOH ( v/v ) for 20 h at 4°C . After separation from the aqueous phase following the addition of 20% H2O ( w/v ) , the organic phase was dried . The resulting material was resuspended in ice-cold acetone and incubated for 20 h at −20°C . The acetone-soluble fraction was then dried , resuspended in ethanol , and loaded onto a silica gel TLC plate and eluted with 90/9/1 CHCl3/MeOH/H2O as the mobile phase . The yellow band corresponding to mycolactone ( retention factor of 0 . 2 ) was then scraped and mycolactone eluted from silica particles using 2/1 CHCl3/MeOH ( v/v ) . Following solvent evaporation , purified mycolactone was resuspended in ethanol . The concentration of the resulting solution was determined by UV absorption , as described [13] . As recently described [13] , mycolactone or lipid extracts were applied to a silica gel TLC plate and eluted with 90/9/1 CHCl3/MeOH/H2O as the mobile phase . The eluted TLC plate was briefly warmed on a hot plate to evaporate the organic solvents , and quickly immersed into a 0 . 1 M acetone solution of 2-naphtalene boronic acid ( Sigma ) , then heated to 100°C for 5∼10 seconds . The TLC plate was then irradiated with a UV reader equipped with a 312 nm lamp . Mycolactone or lipid extracts were also analyzed by High Performance Liquid Chromatography coupled to Mass Spectrometry ( HPLC/MS/MS ) . We used a Shimadzu HPLC fitted with a BDS Hypersil C8 column ( 5 µm , 4 . 6×250 mm ) , with UV detection at 360 nm . Mycolactone was eluted by a 60 min gradient from 50 to 95% acetonitrile in water after 22 min . Fractions of interest were collected in glass tubes and analysed on a QSTAR XL , AB-MDS-SCIEX mass spectrometer with an electrospray ion source and the following parameters: ion spray voltage ( IS ) , 5200 v; curtain gas ( CUR ) , 25; gas 1 ( GS1 ) , 5; declustering potential ( DP ) , 50 v; focusing potential ( FP ) , 225 v; declustering potential 2 ( DP2 ) , 15 v . Mycolactone was identified by the presence of [M+Na]+ m/z 765 . 5 . MS-MS parameters were: ion spray voltage ( IS ) , 6000 v; curtain gas ( CUR ) , 25; gas 1 ( GS1 ) , 20; declustering potential ( DP ) , 60 v; focusing potential ( FP ) , 230 v; declustering potential 2 ( DP2 ) , 10 v; collision energy ( CE ) , 60; collision gaz ( CAD ) , 5 . Data were collected and processed through the Analyst QS 1 . 1 software from AB-MDS-Sciex . Wound swabs obtained from the undermined edges of ulcerative lesions were soaked into 1 ml ethanol immediately after sampling and stored in polypropylene collecting tubes at -20°C , protected from light , until analysis . Ethanol solutions of exudates were concentrated to a volume of 500 µl then processed to TLC-Fluo and HPLC/MS/MS analyses . Lipids were extracted from serum samples ( 1 ml ) by sequential addition of 4/1 MeOH ( v/v ) , 1/1 CHCl3 ( v/v ) , and 3/1 H2O ( v/v ) , each step being followed by thorough mixing . The upper aqueous phase was discarded and the bottom organic phase transferred to a glass tube containing 3 ml of MeOH . The resulting solution was centrifuged to sediment insoluble particulate matter . The soluble organic phase was then dried and the resulting product re-suspended in ethanol for TLC-Fluo and HPLC/MS/MS analyses . Mononuclear cells were isolated from 10 ml whole blood by differential sedimentation on Ficoll-Hypaque ( GE Healthcare ) . Cell pellets were then dried and stored at −20°C until lipid analysis . Total lipids were extracted from cell pellets by addition of 1 ml 2/1 CHCl3/MeOH ( v/v ) for 48 h at 4°C . The organic phase was recovered by addition of 20% H2O ( w/v ) , dried and the resulting product resuspended in ethanol for TLC-Fluo and HPLC/MS/MS analyses .
We first compared the TLC-Fluo and HPLC/MS/MS approaches for mycolactone detection in the 0–500 ng range , using purified mycolactone as a reference . Both methods were highly sensitive , yielding a detectable signal with only 10 ng mycolactone ( Fig . 1A , B ) . However the fluorescent signal of mycolactone modified by coupling to 2-naphtalene boronic acid was not a linear function of its concentration , whereas the areas of mycolactone elution peaks in HPLC were proportional to mycolactone concentration in the 10–500 ng domain ( Fig . 1C ) . To evaluate the efficacy of mycolactone extraction from biological samples , serum samples ( 1 ml ) or mononuclear cell pellets ( 106 cells ) were spiked with purified mycolactone . We then proceeded to solvent extraction as described in the Methods section , and estimated the proportion of recovered mycolactone using the above-described standard curves . The estimated yields of extraction with organic solvents were of 20% from cell pellets and 10% from serum samples ( not shown ) . Several combinations of solvents were tried that did not improve the recovery of mycolactone . Exudate samples were split into two equivalent parts , which were analyzed in parallel for the presence of mycolactone by HPLC/MS/MS or TLC-Fluo . We could not conclude on the presence of mycolactone by TLC-Fluo , because of the co-migration of auto-fluorescent compounds ( Fig . 2A ) . Using HPLC determination , elution peaks corresponding to mycolactone were observed in 3/6 newly diagnosed patients and all patients undergoing or completing their course of antibiotic treatment ( 13/13 , 4/4 respectively ) ( Table 1 ) . To confirm that they effectively contained mycolactone , elution peaks were collected in three patients , and analyzed by MS/MS . In all of them , the characteristic spectrum of mycolactone parent ion ( m/z 765 ) and products was observed [14] , demonstrating the presence and structural integrity of mycolactone in ulcer exudates ( Fig . 2B ) . Notably , the presence of mycolactone in these samples persisted during and after completion of antibiotic therapy ( Fig . 3 ) . Since our previous experiments in experimentally infected mice demonstrated the presence of mycolactone in circulating mononuclear cells , we analyzed this biological material in BU patients . Whole blood ( 10 ml ) was collected in patients from Ghana and Ivory Coast at various stages of the disease ( Table 1 ) and mononuclear cells were isolated by Ficoll gradient centrifugation . However , no mycolactone could be identified in any of the samples tested ( n = 52 ) by TLC-Fluo or HPLC/MS/MS ( not shown ) . Total lipids were extracted from 1 ml serum samples then analyzed by TLC-Fluo ( Ghana samples ) or HPLC ( Ivory Coast samples ) ( Table 1 ) . Again , the identification of mycolactone by TLC-Fluo was difficult , because of the co-migration of auto-fluorescent compounds ( Fig . 4A ) . Using HPLC , mycolactone detection was also hampered by the co-elution of UV-absorbing contaminants ( Fig . 4B ) . We could nevertheless identify mycolactone-like peaks in 3/5 newly diagnosed patients , 1/8 patients undergoing treatment , and 0/4 patients completing treatment . These peaks were collected in two patients and analyzed by MS/MS . In both of them , the characteristic spectrum of mycolactone parent and product ions could be observed ( Fig . 4B ) . The concentration of circulating mycolactone was evaluated in positive samples by measuring the area of mycolactone elution peaks and subtracting the mean signal of two healthy controls ( Fig . 5 ) . Calculated values were in the 40–200 ng/ml range .
In the present study , we investigated whether mycolactone was detectable in easily accessible samples of BU patients . We used two chemical approaches , both requiring the extraction of total lipids by organic solvents . The efficacy of this extraction step , as measured by addition of pure mycolactone to control samples , was mediocre and reduced dramatically the sensitivity of the following TLC-Fluo and HPLC determinations . An explanation for such a limited yield of extraction may be that mycolactone associates with biomolecules preventing solvent access , a possibility that we are currently testing by studying the impact of various thermal and enzymatic treatments . If the efficacy and selectivity of mycolactone extraction can be improved , the recently described and field-friendly TLC-Fluo detection method may still be an option for mycolactone-based point-of-care diagnostic tests . If not , it will be necessary to design alternative approaches that do not require this purification step . Using HPLC/MS/MS , we could demonstrate that mycolactone gains access to the peripheral blood of human patients . In previous studies in the mouse model , we detected structurally intact mycolactone in mononuclear cell fractions of pooled blood samples harvested from mice subcutaneously injected with mycolactone , or experimentally infected with M . ulcerans [12] . In the present work , mycolactone could not be identified in blood mononuclear cells . This cell subpopulation was isolated from 10 ml whole blood , and we estimated the maximum yield of mycolactone extraction from mononuclear cell pellets to 20% . If mycolactone effectively reaches blood mononuclear cells in human patients , its cellular concentration may be too low to be detected in the accessible volume of blood . Alternatively , mycolactone may be unstable in the conditions used in the present study to isolate and store mononuclear cell pellets . In contrast , we were able to demonstrate the presence of structurally intact mycolactone in the serum of 3/5 newly diagnosed BU patients ( Table 1 ) . This novel information provides the essential proof of concept for the design of BU diagnostic tests based on mycolactone detection in peripheral blood . Since mycolactone is extracted from serum samples with low efficacy , the number of positive samples and the calculated concentrations of circulating mycolactone are probably underestimated in this study . Whether mycolactone kinetics in serum could be employed to monitor the response to antibiotic treatment will certainly be interesting to investigate further . Our preliminary results suggest that , in spite of a sustained presence in ulcer exudates ( Fig . 3 ) , mycolactone concentration ( Fig . 5 ) showed a tendency to decrease in the serum during antibiotic therapy . If confirmed by longitudinal studies , the decay of circulating mycolactone during antibiotic therapy would provide an explanation to the recovery of cellular immune responses during treatment [15] , [16] and after surgical excision of BU lesions [17] . Here we considered patients at ulcerative stages of the disease . BU is usually diagnosed on the basis of clinical symptoms , as the identification of M . ulcerans by means of cultures or PCR requires dedicated facilities and specialized equipment ( reviewed in [18] ) . Common differential diagnoses of BU include other tropical ulcers ( venous , phagedenic , neurogenic ) , leishmaniasis , yaws and squamous cell carcinoma . The presence of biologically active mycolactone was recently demonstrated in skin biopsies of BU patients [11] . Here we show that mycolactone can be detected in ulcer exudates obtained non-invasively from wound swabs , in a structurally intact form and at concentrations in the 50–200 ng/ml range , which strongly suggests that mycolactone detection in exudates may be of interest for differential diagnosis of the ulcerative forms . Whether mycolactone is present at pre-ulcerative stages in fine-needle aspirates and in the peripheral blood is currently under investigation . We observed significant amounts of mycolactone in exudates at the end of antibiotic therapy . In some instances , antibiotic pressure may not efficiently block , and could even enhance the production of a toxin . For instance , in patients infected with E . coli O157∶H7 , the use of antimicrobials has been discouraged because it stimulates toxin production and augments the risk of detrimental , or even fatal complications [19] . Whether antibiotics promote the expression of mycolactone by M . ulcerans is unknown . However , their efficacy at killing M . ulcerans is not questionable with 4 weeks of treatment leading to culture negativity [3] , [4] , [5] , [6] , [7] . Although a stimulatory effect of antibiotics on mycolactone production cannot be excluded , our observation suggests that mycolactone persists in cutaneous tissues after the demise of M . ulcerans . Since mycolactone displays inherent ulcerative properties [8] , this phenomenon may explain why some BU take considerable time to heal despite remaining culture negative . The administration of antibiotics is sometimes associated with adverse reactions , excessive inflammation and pain sensation , that paradoxically make the symptoms of infection worse [20] . It is possible that in BU , like in the Jarish-Heixheimer reaction in syphilis or in relapsing fevers , efficient killing of M . ulcerans by antibiotics leads to the sudden and massive release of bacterial antigens locally that act as immuno-stimulants . Mycolactone displays potent immunosuppressive properties in vitro that are thought to contribute to the cellular response defects of BU patients [21] , [22] , [23] , [24] , [25] . The rapid decline of mycolactone from the systemic circulation during treatment may thus provoke exuberant inflammatory responses at the level of the lesions , and cause the paradoxical reactions observed in BU disease [26] . | Mycolactone is a diffusible cytotoxin produced by Mycobacterium ulcerans , the causative agent of the skin disease Buruli ulcer . In a previous study using animal models , we reported that mycolactone released by cutaneous foci of infection gains access to the peripheral blood . Here we investigated whether mycolactone circulates in human patients and is detectable in clinically accessible samples . Using a combination of solvent extraction , high performance liquid chromatography and mass spectrometry analysis , we found that structurally intact mycolactone was present in ulcer exudates obtained from wound swabs and in the serum of patients . Unexpectedly , high titres of mycolactone were detected in ulcer exudates after completion of antibiotic treatment . In contrast , mycolactone could only be detected in the serum of newly diagnosed patients . Our results demonstrate that mycolactone detection in serum and ulcer exudates may be used to diagnose BU . Moreover , they suggest that the kinetics of mycolactone concentration in serum may be indicative of the clinical response of patients to antibiotic therapy . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine"
] | 2011 | Mycolactone Diffuses into the Peripheral Blood of Buruli Ulcer Patients - Implications for Diagnosis and Disease Monitoring |
Efforts to control the spread of Buruli ulcer – an emerging ulcerative skin infection caused by Mycobacterium ulcerans - have been hampered by our poor understanding of reservoirs and transmission . To help address this issue , we compared whole genomes from 18 clinical M . ulcerans isolates from a 30km2 region within the Asante Akim North District , Ashanti region , Ghana , with 15 other M . ulcerans isolates from elsewhere in Ghana and the surrounding countries of Ivory Coast , Togo , Benin and Nigeria . Contrary to our expectations of finding minor DNA sequence variations among isolates representing a single M . ulcerans circulating genotype , we found instead two distinct genotypes . One genotype was closely related to isolates from neighbouring regions of Amansie West and Densu , consistent with the predicted local endemic clone , but the second genotype ( separated by 138 single nucleotide polymorphisms [SNPs] from other Ghanaian strains ) most closely matched M . ulcerans from Nigeria , suggesting another introduction of M . ulcerans to Ghana , perhaps from that country . Both the exotic genotype and the local Ghanaian genotype displayed highly restricted intra-strain genetic variation , with less than 50 SNP differences across a 5 . 2Mbp core genome within each genotype . Interestingly , there was no discernible spatial clustering of genotypes at the local village scale . Interviews revealed no obvious epidemiological links among BU patients who had been infected with identical M . ulcerans genotypes but lived in geographically separate villages . We conclude that M . ulcerans is spread widely across the region , with multiple genotypes present in any one area . These data give us new perspectives on the behaviour of possible reservoirs and subsequent transmission mechanisms of M . ulcerans . These observations also show for the first time that M . ulcerans can be mobilized , introduced to a new area and then spread within a population . Potential reservoirs of M . ulcerans thus might include humans , or perhaps M . ulcerans-infected animals such as livestock that move regularly between countries .
Buruli ulcer ( BU ) is a neglected tropical disease caused by infection with Mycobacterium ulcerans . Each year 5000–6000 cases are reported from 15 of the 33 countries where BU cases have been reported , predominantly from rural regions across West and Central Africa [1] . The disease involves subcutaneous tissue and has several manifestations but necrotic skin ulcers are a common presentation , caused by the proliferation of bacteria beneath the dermis by virtue of a secreted bioactive lipid called mycolactone [2] . The role of mycolactone in the natural ecology of M . ulcerans is not understood , but it has been shown to possess several specific activities against mammalian cells from activating actin polymerization , blocking secreted protein translocation , to interacting with neuronal angiotensin type II receptors causing hypoesthesia [3–5] . These collective biological activities of mycolactone , while diverse , might collectively help explain the tissue destruction , lack of inflammation , and painlessness associated with BU . BU is rarely fatal and early diagnosis followed by combined antibiotic therapy ( rifampicin and streptomycin ) is key to preventing complications that can arise from severe skin ulceration [6] . Epidemiological studies frequently link BU occurrence with low-lying and wetland areas and human-to-human transmission seems rare , suggesting an environmental source of the mycobacterium [7–23] . Frustratingly however , the environmental reservoir ( s ) and mode ( s ) of transmission of M . ulcerans remain unknown . M . ulcerans has the genomic signature of a niche-adapted mycobacterium , indicating that it is unlikely to be found free-living in diverse aquatic ( or other ) environments , but more likely in close association with a host organism . In south eastern Australia , native marsupials have been identified as both susceptible hosts and reservoirs of M . ulcerans , with high numbers of the bacteria shed in the feces of infected animals . Mosquitoes have also been found to harbor the bacteria in this region and a zoonotic model of disease transmission has been proposed involving possums , biting insects and humans [24–26] . No such animal reservoir has yet been identified in African BU endemic areas and studies of BU lesion distribution are thought not consistent with mosquito biting patterns [22 , 27] . On the other hand , case-control studies in Cameroon have shown that bed nets are protective , supporting a role for insects in transmission [28] . A feature of M . ulcerans is the close correlation between genotype and the geographic origin of a strain , but its restricted genetic diversity has limited the application of traditional molecular epidemiological methods such as VNTR-typing to discriminate between isolates at the village or even regional scales . The advent of low cost genomics has opened up new possibilities to explore and track the movement and spread of this pathogen within communities [29 , 30] . Agogo is the principal town of 30 , 000 inhabitants in the Asante Akim North ( AAN ) district within the Ashanti region of Ghana and BU has been reported in about half of the sixty-four communities in this district since mid-1975 [10] . The AAN district covers an area of 650 km2 in the forest belt of Ghana and it is the third most endemic district in Ghana [31] . Five of the communities ( Ananekrom , Serebouso , Nshyieso , Serebuoso and Dukusen ) in this district are among the communities reported with the highest burden of the disease in Ghana [31] . About 120 laboratory-confirmed new cases are reported annually in this district [31] . Subsistence farming and petty trading are the principal occupations of inhabitants of these endemic communities . People generally live in simple dwellings constructed from local materials . Houses are often close together with 3–5 households in a compound . Many inhabitants raise animals such as goats , sheep , and pigs in the immediate vicinity of their houses . Farming is the main occupation with some people engaged in fishing and petty trading . Farms may be distant , ranging 5–20 km from a given domicile . Fishing is usually undertaken close to home . Water sources are of two types . Water for drinking and cooking is usually fetched from bore holes fitted with mechanical pumps , within or near a village . Water for bathing and domestic chores such as washing of clothes is drawn from local natural water sources ( rivers , streams , ponds ) . These natural sources are usually no more than 500 metres from a given village . In this study we sequenced and compared the genomes of 18 M . ulcerans isolates obtained from 10 BU endemic villages in the AAN district and uncovered genetic evidence supporting the introduction of a foreign clone of M . ulcerans to this region . This observation indicates that M . ulcerans can be mobilized and spread throughout a region , indicating that reservoirs of the bacterium are themselves potentially highly mobile .
M . ulcerans isolates were obtained from BU diagnostic samples , collected as part of routine laboratory diagnosis . Ethical approval to interview patients and use bacterial isolates resulting from diagnostic specimens for research was obtained from the ethical review board of the Noguchi Memorial Institute for Medical Research , University of Ghana , Legon , Accra , Ghana ( FWA 00001824 ) , with written informed consent obtained from all adult patients or the parents/guardians of the participating children . The study was carried out in ten endemic villages including Ananekrom , Nshyieso , Serebouso , Dukusen , Afreserie , Afreserie OK , Baama , Nysonyameye , Kwame Addo and Bebuso , in the Asante Akim North ( AAN ) district of Ghana ( Table 1 ) . These are small villages and hamlets , 5 to 10 km from each other with populations between 120–1500 inhabitants . Ananekrom is the largest of these communities and is the closest ( 15 km ) to the district capital , Agogo . An asphalt road connects Agogo to Ananekrom , Dukusen and Afriserie , while the other communities are located off this main road and are connected to each other by unmade roads and foot-tracks . A community health centre Ananefromh ( near Ananekrom ) is usually the first point of call for patients seeking medical treatment . Patients suspected of having BU are referred to the Agogo Presbyterian Hospital for diagnosis and treatment . Patient information including name and place of residence were obtained from hospital records and patients were visited in their homes for more detailed interviews that included questions about possible travel to other BU endemic areas outside the AAN district . GPS coordinates in the vicinity of each patient’s residence were recorded in order to map the spatial distribution of cases in the villages , based on the assumption that the patient acquired their infection near their domicile . The isolates examined in this study are listed in Table 1 and were recovered from fine needle aspirates ( FNA ) or swabs , obtained from pre-ulcerative lesions and ulcers respectively . Specimens were stored in transport medium and PBS and transported in cool boxes to the Noguchi Memorial Institute for Medical Research ( NMIMR ) for diagnosis [32 , 33] . Tubes containing swabs were vortexed in 3 ml of transport medium for 30 sec and the swabs removed . A volume of 250μl of the transport medium from either specimen type was transferred into 1 . 5 ml microfuge tubes and decontaminated using the oxalic acid method as previously described [34] . The pellets were resuspended in 100 μl phosphate buffered-saline ( PBS ) and 100 μl volume of the decontaminated sample was inoculated onto Löwenstein Jensen ( LJ ) slopes and incubated at 33°C . The cultures were observed weekly for growth . Suspected M . ulcerans colonies were harvested and DNA extracted as described above [35] . The DNA extract was tested with the IS2404 PCR for the identification of M . ulcerans [36] . Colonies positive for IS2404 were suspended in 1 ml of Middlebrook 7H9 broth and stored at -80°C . All 18 bacterial samples analyzed were selected from this stored collection and were subcultured on LJ medium and DNA for whole genome sequencing was extracted from resulting growth as described [35] . The isolation date refers to the date when colonies became visible on LJ medium following primary cultivation . DNA sequencing was performed using two methods . The Ion Torrent Personal Genome Machine was employed , with a 316 chip and 200bp single-end sequencing chemistry ( Life Technologies ) . Genomic libraries for Ion Torrent sequencing were prepared using Ion Express , with size selection using the Pippin Prep ( Sage Sciences ) and emulsion PCR run using a One-Touch instrument ( Life Technologies ) . The Illumina MiSeq was also used , with Nextera XP library preparation and 2x250 bp sequencing chemistry . Read data for the study isolates have been deposited in the European Nucleotide Archive ( ENA ) under accession ERA401876 . Prior to further analysis , reads were filtered to remove those containing ambiguous base calls , any reads less than 50 nucleotides in length , and containing only homopolymers . All reads were furthermore trimmed removing residual ligated Nextera adaptors and low quality bases ( less than Q10 ) at the 3' end . Resulting sequence Fastq sequence read files from either platform were subjected to read-mapping to the M . ulcerans Agy99 reference genome ( Genbank accession number CP000325 ) using Bowtie2 v2 . 1 . 0 [37] with default parameters and consensus calling to identify SNPs ( indels excluded ) using Nesoni v0 . 109 , a Python utility that uses the reads from each genome aligned to the core genome to construct a tally of putative differences at each nucleotide position ( including substitutions , insertions , and deletions ) ( www . bioinformatics . net . au ) . Those positions in the Agy99 reference genome that were covered by at least 3 reads from every isolate defined a core genome . Note that the pMUM001 plasmid ( required for mycolactone synthesis ) was not included in the reference genome [38] . Testing of the plasmid sequences revealed less then 10 polymorphic sites among the genomes under investigation and the highly repetitive sequence structure of the mycolactone genes impaired unambiguous read-mapping . An unpaired t test with Welch’s correction was used to assess the differences between mean nucleotide pairwise identities for different groups of genomes . The null hypothesis ( no difference between means ) was rejected for p<0 . 01 . The inputs for subsequent phylogenomic analyses were the nucleotide sequence alignments of the concatenated variable nucleotide positions for the core genome among all isolates . A maximum-likelihood ( ML ) phylogeny was inferred using RAxML v 7 . 2 . 8 , with the GTR model of nucleotide substitution ( plugin within Geneious v 8 . 0 . 4 ) . We performed 1000 rapid pseudo-replicate bootstrap analyses to assess support for the ML phylogeny . We used Consensus-Tree-Builder ( Geneious v8 . 0 . 4 ) to collapse nodes in the tree with bootstrap values below a set threshold of 70% . The resulting phylogenomic tree was exported in Newick format and visualized using FigTree v1 . 4 . 0 ( tree . bio . ed . ac . uk/software/figtree ) . A haplotype network was derived using the median-joining algorithm as implemented in SplitsTree v4 . 13 . 1 [39 , 40] . A correction to the source attribution of the M . ulcerans Agy99 reference genome was also made in the course of this study , where it was realized that this isolate was actually obtained from a patient attending St Martin’s Hospital in Agroyesum ( Amansie West ) and not the Ga District Hospital as originally published [41] ( K . Asiedu and J . , Hayman pers comms ) , thereby explaining the inconsistent geographic clustering reported in previous molecular epidemiological studies [30 , 42 , 43] .
Eighteen M . ulcerans isolates were randomly selected for whole genome sequencing . The isolates represented 20% ( total of 92 isolates from 2010–2012 ) of all culture-confirmed BU cases referred to the Agogo Presbyterian Hospital between 2010 and 2012 ( Table 1 ) . There were no differences in colony phenotype or growth characteristics among the isolates . The DNA sequence reads for each genome were mapped to the M . ulcerans Agy99 reference sequence . Sequencing and read-mapping summary statistics are given in Table 1 . In addition to the 18 Agogo isolates sequenced here , 15 other genomes ( including some previously described ) were included in comparisons making a total of 33 isolates ( Table 1 ) . These additional genomes were from M . ulcerans isolates in other regions of Ghana and from surrounding countries to provide appropriate genetic context for interpreting the diversity and evolution of M . ulcerans isolates from around Agogo . Read-mapping and SNP identification revealed 320 variable nucleotide positions across a 5 . 2Mb core genome for the 33 isolates . A phylogeny was inferred from this alignment , showing the clustering typical of M . ulcerans genotypes with geographic origin ( Fig . 1 ) . A separate SNP alignment was performed taking the genome sequences for only the 18 isolates from the Agogo region , and 10 of them ( called Agogo-1 ) clustered with isolates from the neighboring district of Amansie West and also the Ivory Coast , the country which borders this region to the west ( Fig . 1 ) . This close relationship is indicative of a local clone that has spread and persisted within the greater region for some time . Unexpectedly however , this analysis also revealed the presence of a second distinct M . ulcerans genotype co-circulating with Agogo-1 . This second genotype ( called Agogo-2 ) was substantially more diverse from all other Ghanaian M . ulcerans genotypes ( 138 SNPs ) , suggesting the re-introduction of M . ulcerans to the Agogo region , potentially from a source outside Ghana ( Fig . 1 , S1 Table ) . The intra-genotype variation within either cluster was low . The mean nucleotide pairwise identity was 94 . 7% ( SEM ± 0 . 4 ) for Agogo-1 versus 97 . 2% ( SEM ± 0 . 4 ) for Agogo 2 . The mean pairwise nucleotide identity was significantly lower for Agogo-2 genomes compared with Agogo-1 ( p<0 . 001 ) . To investigate the possible origin of the Agogo-2 isolates we compared SNP profiles among our panel of M . ulcerans genomes from across West and Central Africa . The closest match obtained was to isolate ITM102686 , obtained from a patient originating from Ibadan , Nigeria , with 29 SNPs different when only this genome was compared to the Agogo-2 cluster . This close association may indicate that Nigeria was the source of the Agogo-2 cluster . Some circumspection is needed when interpreting these data , as only two M . ulcerans genomes were sampled from countries east of Benin . There is however a compelling patient history behind this isolate to support Nigeria as the correct origin . The Caucasian patient , a long-term resident in Ibadan and an employee of a non-government organisation , believes he was infected on an Ibadan golf course , when he was bitten by black biting flies ( his description suggests they may have been moth flies [Psychodidae] ) that began plaguing the course when ground works started adjacent to a lake on the course . The patient developed a painful ulcer on the site of the insect bites . A couple of months later he developed a second ulcer on an adjacent site on the same limb that was microbiologically diagnosed as a Buruli ulcer . This patient history , combined with the documented cases of BU in Ibadan , with cases occurring around the Ibadan University campus and other nearby institutions [44] , support Ibadan as the likely origin of M . ulcerans isolate ITM102686 . We next explored the distribution of M . ulcerans genotypes in the Agogo region at the village scale and observed no obvious pattern or relationship between genotype , patient , strain and village ( Fig . 2 ) . There is complete intermixing of Agogo-1 and Agogo-2 clusters amongst the population . Median-joining-network analysis suggested the independent radiation of the two clusters throughout the region ( Fig . 2 ) . Furthermore , within either cluster there was a broad distribution of cluster subtypes across the region . For example , isolates F70 and S38 ( Agogo-1 ) have identical SNP profiles but the patients came from Baama and Serebouso , villages separated by 10–15 km . Similarly isolates F74 and 1510 ( Agogo-2 ) , came from patients who live in two different villages ( Fig . 2 ) . Patient interviews did not identify any travel histories or other epidemiological links that might explain these distribution patterns . An 11-year old girl from Serebouso was the third child within her family to have BU ( isolate 212 , November 2012 ) , eight years after two of her siblings had the disease . The family of this child lived very close to that of another BU patient , a 3 year-old infant ( isolate S77 , February 2010 ) . Both isolates belonged to the Agogo-2 cluster but their genome sequence differed in nine nucleotide positions , a significant amount of genetic variation given that S77 shares a near identical genotype with F74 and 1510 . Again , we could not identify any specific activity or travel history such as attending a common community event that was shared by Agogo-2 genotype patients . These data suggest that ( i ) the disease is acquired locally , ( ii ) multiple M . ulcerans genotypes are circulating simultaneously within the local region and ( iii ) a single clone can have the propensity to spread through a region . Further support for local acquisition of infection comes from observations of infants with no travel history with BU such as a locally-born 2-year old infant from Ananekrom identified over the time of this study .
The clonal population structure of M . ulcerans has made identifying and comparing genetic variation in isolates at anything less than a continental scale very difficult . Here we have used the high resolution afforded by comparative genomics to explore the molecular epidemiology of BU at the regional and village scale . Like recent studies using a single polymorphic genetic locus or whole genome sequence comparisons to assess M . ulcerans genetic diversity across a range of African countries , we found a highly significant relationship between the genotype of an isolate and its geographic origin at a national and regional scale [42 , 45] . These repeated observations indicate that M . ulcerans , when introduced to an area , remains localized and isolated for a sufficiently long period to allow mutations to become fixed in the bacterial population and a local genotype to evolve . It is reasonable to infer therefore , that the environmental reservoirs of the bacterium in these areas are also likely to be somewhat localized and isolated . However , the current study has shown for the first time how this focal distribution pattern breaks down at a local scale with the presence of identical genotypes appearing concurrently in separate areas of the same district . There was no discernible distribution pattern for either the Agogo-1 or Agogo-2 genetic clusters , with both M . ulcerans genotypes appearing at the same times and within the same villages across the region . Interestingly , there were several examples of isolates with identical genome sequences ( e . g . isolates F74 , 1510 or F85 , F65 ) that were obtained from patients living in four different villages , each separated by distances in excess of 10km ( Fig . 2 ) . There are several potential explanations for these patterns . The bacteria ( or a vector spreading the bacteria ) may be widely distributed across the region and infections are being acquired locally , or it may be that people are traveling and becoming infected from a common point source . Patient interviews and travel histories did not reveal any common activity that might explain a point-source transmission scenario , although the long incubation time for this disease ( 4-months ) is likely to make recall of any such events unreliable [46] . However , on balance the former scenario seems most likely , and we suggest that each genotype of M . ulcerans has now spread equally widely across the region . If this assumption is correct , then the lack of genetic variation among isolates suggests that the spread of M . ulcerans throughout the region has occurred relatively rapidly , with insufficient time elapsed for mutations to accumulate . Reliable mutation rates for M . ulcerans have not been established and some solid data here would allow inferences regarding the time particular clones have been extant within a population . To our knowledge , this is the first report to employ whole genome sequencing to explore the molecular epidemiology of BU at a local scale . A previous study utilizing high-resolution SNP assays to explore M . ulcerans genetic variation did uncover some suggestion of local genotype clustering and a recent report used VNTR to examine the link between human and environmental sources of M . ulcerans [30 , 47] . However such approaches rely on variable nucleotides that have been defined from a limited reference genome set . If this reference genome set does not represent the genetic variation of the isolates under investigation then data analysis can be flawed , with phenomena such as long-branch attraction and phylogenetic discovery bias confounding analyses [48] . Whole genome sequencing and comparisons of all isolates under investigation as in our study here overcomes the potential weaknesses of targeted SNP-based typing . SNP-typing could however be employed to classify patient samples as Agogo-1 and Agogo-2 genotypes without relying on sequencing of cultured isolates , as culture sensitivity is only around 30% , depending on transport duration . Future studies could thus search for clinical phenotypes between these two distinct bacterial genotypes , although no differences were observed in pathology or treatment outcomes among the patients associated with this study . There are interesting parallels between M . ulcerans and Mycobacterium leprae , the causative agent of leprosy , where genomics has shown that the leprosy bacillus is another example of a niche-adapted , highly clonal , zoonotic mycobacterial pathogen , with the potential to spread from environment-to-human [49–52] . Mycobacterium tuberculosis might also be considered in a similar context , with genomic population analysis also suggesting interactions among genetically distinct M . tuberculosis lineages [53 , 54] . One potential issue arising from this study is the risk of incorrectly attributing Nigeria as the origin M . ulcerans genome sequence ITM102686 , as it represents only one isolate . While the patient history makes a persuasive argument for Ibadan as the source of the infection , additional M . ulcerans isolates are clearly required from patients in different BU endemic regions of Nigeria and surrounding countries , to further explore the relationship and disease transmission patterns we propose here . Regardless of the precise origin of Agogo-2 isolates , the data presented here suggest that M . ulcerans can be introduced into a region and then be spread extensively . How might M . ulcerans be imported into a region ? We speculate that movements of people or perhaps animals between countries could be one likely means , where infected individuals with BU lesions that can contain very high bacterial burdens might inadvertently contaminate aquatic environments during bathing or other water contact activities . Now is the time to undertake more intensive and extensive whole-genome M . ulcerans sequencing surveys across West Africa , to assess the extent of genotype admixture such as we’ve revealed here . Enriching our genome data will also inform other research programs that are identifying reservoirs of M . ulcerans , leading to the new knowledge required to design interventions and stop the spread of BU . | In this study we use the power of whole genome sequence comparisons to track the spread of Mycobacterium ulcerans , the causative agent of Buruli ulcer , through several villages in the Ashanti region of Ghana , providing new insights on the behaviour of this enigmatic and emerging pathogen . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Whole Genome Comparisons Suggest Random Distribution of Mycobacterium ulcerans Genotypes in a Buruli Ulcer Endemic Region of Ghana |
Immune senescence , defined as the age-associated dysregulation and dysfunction of the immune system , is characterised by impaired protective immunity and decreased efficacy of vaccines . Recent clinical , epidemiological and immunological studies suggest that Cytomegalovirus ( CMV ) infection may be associated with accelerated immune senescence , possibly by restricting the naïve T cell repertoire . However , direct evidence whether and how CMV-infection is implicated in immune senescence is still lacking . In this study , we have investigated whether latent mouse CMV ( MCMV ) infection with or without thymectomy ( Tx ) alters antiviral immunity of young and aged mice . After infection with lymphocytic choriomeningitis virus ( LCMV ) or Vaccinia virus , specific antiviral T cell responses were significantly reduced in old , old MCMV-infected and/or Tx mice compared to young mice . Importantly , control of LCMV replication was more profoundly impaired in aged MCMV-infected mice compared to age-matched MCMV-naïve or young mice . In addition , latent MCMV infection was associated with slightly reduced vaccination efficacy in old Tx mice . In contrast to the prevailing hypothesis of a CMV-mediated restriction of the naïve T cell repertoire , we found similar naïve T cell numbers in MCMV-infected and non-infected mice , whereas ageing and Tx clearly reduced the naïve T cell pool . Instead , MCMV-infection expanded the total CD8+ T cell pool by a massive accumulation of effector memory T cells . Based on these results , we propose a new model of increased competition between CMV-specific memory T cells and any ‘de novo’ immune response in aged individuals . In summary , our results directly demonstrate in a mouse model that latent CMV-infection impairs immunity in old age and propagates immune senescence .
Immune senescence , defined as the age-related alterations of the immune system , is associated with an increased incidence of infections , cancer , autoimmunity and a reduced efficacy of prophylactic vaccines [1]–[3] . Although all components of the immune system undergo age-related changes , the T cell compartment is most significantly affected by a quantitative and qualitative loss of naïve T cell diversity due to declining thymic output and increasing dysregulation of compensatory homeostatic mechanisms [4]–[8] . Therefore , ageing hosts have increasing difficulties to mount efficient primary T cell responses whereas memory maintenance and recall responses appear to be less affected [9]–[11] . Immune senescence is certainly a multifactorial process involving genetic , molecular , cellular and also environmental factors . Among the latter , Cytomegalovirus ( CMV ) infection has gained considerable interest in recent years as a potential propagator of human immune senescence [12] . First , two independent epidemiological studies have linked human ( H ) CMV-seropositivity with decreased overall survival of elderly [13] , [14] . Second , a series of Swedish cohort studies of very elderly have identified a so called ‘immune risk profile’ ( IRP ) , which was strongly predictive of all cause mortality . Importantly , HCMV-infection was one of the most important IRP-parameters [15] . Third , protective antibody titres after influenza vaccination were reduced in HCMV-positive compared to HCMV-negative elderly individuals , although this finding was not confirmed in a subsequent study [16] , [17] . Together , these studies suggest that HCMV-infection may be associated with decreased immunocompetence of the elderly . However , it is unclear whether HCMV-infection is causally linked with accelerated immune senescence or whether it is just a marker for something else - like poor nutrition during childhood - since HCMV-infection is known to be associated with lower socio-economic resources [18] . Moreover , very little solid information is available about potential underlying mechanisms of HCMV-induced immune senescence . Immunologically , HCMV-infection is characterised by inducing very prominent T cell responses with the highest magnitude of all investigated persistent pathogens , and this response regularly occupies 20% and more of the total CD8+ T cell pool [19] . In addition , T cell responses seem to increase with duration of infection due to memory inflation both in mice and in humans [20]–[25] . This has fostered the hypothesis that these massive CMV-driven memory T cell expansions significantly accelerate the age-associated loss of naïve T cells which are indispensible for potent ‘de novo’ immune responses [26]–[28] . Others have speculated that HCMV-infection and HCMV-specific T cells maintain a proinflammatory cytokine milieu which may be suppressive for immunity in the elderly [29] , [30] . Therefore , an animal model for the direct investigation of CMV-enhanced immune senescence would be extremely informative 1 ) to establish a clear link between CMV-infection and accelerated immune senescence , 2 ) to quantify the influence of CMV , 3 ) to elucidate potential viral and immunological mechanisms , and 4 ) to develop intervention strategies for its prevention . In this study , we have combined mouse ( M ) CMV infection with or without thymectomy ( Tx ) of young adult mice to generate a model for CMV-enhanced immune senescence . By performing Tx in mice we aimed to recapitulate the human situation of a relatively early decline of thymic T cell output [31] . In humans , thymic involution begins shortly after birth and only a minimal number of new thymic emigrants reach the periphery in adults , leading to a progressive loss of naïve T cell numbers and diversity during ageing [32] , [33] . In mice , the thymus remains fully functional for prolonged periods of time in relation to their life expectancy [34] . Our results show that both MCMV-infection and Tx significantly decrease antiviral immune responses and protection of ageing mice . In contrast to Tx , which substantially reduced naïve T cell numbers over time , MCMV-infection did not measurably influence the absolute size of the naïve T cell pool . Instead , MCMV-infection induced a significant and long-lasting expansion of the total CD8+ T cell compartment by the massive accumulation of effector memory T cells ( Tem ) . Therefore , we propose an alternative model of CMV-enhanced immune senescence based on increasing T cell competition . Overall , we present the first data from a mouse model to confirm that latent CMV-infection itself has a propagating influence on the development of immune senescence .
To test whether latent MCMV-infection has an influence on immune control of heterologous viral infections , mice were infected intravenously ( i . V . ) with 107 plaque forming units ( pfu ) MCMV-Δ157 at the age of 6–8 weeks . MCMV-Δ157 is a mutant virus with a targeted deletion of the MCMV open reading frame ( ORF ) m157 encoding a ligand for the NK cell activating receptor Ly49H [35] . After infection of Ly49H-positive C57BL/6 mice with MCMV-Δ157 NK cell stimulation and NK cell mediated early viral control were substantially attenuated [35] . Moreover , infection with MCMV-Δ157 was recently characterised by increased early antigen load , elevated levels of innate cytokines but preserved conventional dendritic cell ( cDC ) function leading to more robust antiviral CD8+ T cell responses which were indispensable for termination of productive MCMV-Δ157 infection [36] . Since we hypothesised that the magnitude of MCMV-specific T cell responses was involved in MCMV-enhanced immune senescence we performed all subsequent experiments using MCMV-Δ157 to enhance the probability of a measurable effect . Despite attenuated NK cell activity , C57BL/6 mice have usually cleared this virus below the limit of detection in all organs within 6–8 weeks indicating the establishment of viral latency ( data not shown ) . Young or old mice with or without latent MCMV-infection were then infected with different , heterologous viruses and viral replication and antiviral CD8+ T cell responses were quantified . First , we analysed viral control and T cell immunity after infection with lymphocytic choriomeningitis virus ( strain WE; LCMV-WE ) , since LCMV elicits a very potent CD8+ T cell response , which is required for early viral control [37] . It has been demonstrated before that aged mice are less efficient in mounting protective LCMV-specific T cell responses [10] . Young ( 4 or 6 months old; 2 or 4 months after MCMV-infection ) and old mice ( 15 or 22 months old; 11 or 18 months after MCMV-infection ) with and without latent MCMV-infection were infected i . v . with 2×103 pfu of the LCMV-WE . Eight days later , LCMV-titres were quantified in different organs ( Figure 1A , B for lung and spleen; liver and kidney not shown ) and the LCMV-specific CD8+ T cell response against two immunodominant epitopes derived from the viral glycoprotein ( GP ) 33 and nucleoprotein ( NP ) 396 was measured in the blood , spleen and lung by tetramer staining or intracellular cytokine staining ( ICS ) ( Figure 1C , D for lung ) . As shown in figure 1A and B , young mice were able to control LCMV-replication within 8 days to levels between 1 . 5–3×103 pfu/organ in spleen and lung , irrespective of latent MCMV-infection . In parallel , total numbers of GP33- and NP396-specific CD8+ T cells were comparable in lung ( Figure 1C , D ) , spleen and blood ( data not shown ) in MCMV-infected and MCMV-naïve young mice . These results indicate that latent MCMV infection does not measurably influence LCMV replication or LCMV-specific CD8+ T cell responses in young mice . As expected , LCMV control was affected by age , since LCMV titres were between 3- and 6-fold higher in old compared to young mice although the difference did not reach statistical significance ( p = not significant ( ns ) ; Figure 1A , B ) . In parallel , total numbers of LCMV-specific CD8+ T cells were between 2- and 3-fold lower in old compared to young mice in all organs tested ( p = ns; Figure 1C , D for lung; spleen and blood not shown ) . These rather small age-associated differences in viral control and LCMV-specific T cell responses are comparable with earlier findings from Kapasi et al [10] . More importantly , LCMV control and LCMV-specific T cell responses were additionally reduced in old MCMV-infected compared to old MCMV-naïve mice . Overall , LCMV titres were between 25- and 1 , 150-fold higher in old MCMV-infected compared to old MCMV-naïve mice in spleen ( 6 . 7×106 vs . 1 . 7×104 pfu/organ , p<0 . 05 ) , lung ( 1 . 1×105 vs . 4 . 5×103 pfu/organ , p<0 . 05 ) , liver ( 5 . 8×107 vs . 5×104 pfu/organ , p<0 . 05 ) and kidney ( 6 . 7×104 vs . 2 . 7×103 pfu/organ , p<0 . 1 ) , and between 40- and 2 , 300-fold higher compared to young mice , independent of MCMV infection ( spleen p<0 . 05 , lung p<0 . 05 , liver , p<0 . 05 , kidney p<0 . 1; Figure 1A , B; data not shown ) . In parallel , GP33- and NP396-specific CD8+ T cell responses were about 15-fold ( lung ) or 3-fold ( spleen ) in old MCMV-infected compared to old MCMV-naïve mice ( p = ns; Figure 1C , D; data not shown ) . Although absolute numbers of LCMV-specific CD8+ T cells were measurably reduced in old mice with or without latent MCMV-infection , we did not identify relevant functional or phenotypic differences between LCMV-specific CD8+ T cells from these four groups of mice with respect to cytokine production ( IFNγ , TNFα ) , cytotoxicity measured by degranulation ( i . e . CD107a-upregulation ) or expression of activation markers like CD43 ( not shown ) . To corroborate these findings in a different viral model , similarly treated four groups of young ( 4 months old ) and old mice ( 25 or 28 months old ) with and without latent MCMV-infection were prepared and infected intraperitoneally ( i . p . ) with 5×106 pfu of Vaccinia virus ( VACV ) recombinant for LCMV-GP ( VACV-GP ) . Six days later , antiviral CD8+ T cell responses were quantified in the blood , spleen and lung and VACV titres were measured in the ovaries . Unfortunately , VACV replication could not be reliably assessed in the ovaries of old and old MCMV-infected mice since we observed massive differences ( up to 105-fold ) of VACV-replication in left and right ovaries of individual mice and within groups of mice . We suspect that VACV-replication in ovaries was heavily influenced by individualised onset of reproductive arrest . However , in young mice , latent MCMV-infection had no measurable influence on VACV-GP clearance ( data not shown ) . Concerning the CD8+ T cell responses , the previous findings using LCMV were confirmed after VACV-infection since we recorded similar trends of attenuated T cell responses in old and latently infected old mice . In young and young MCMV-infected mice , comparable numbers of CD8+ T cells specific for two out of three immunodominant VACV-derived epitopes and for the recombinant LCMV-derived epitope GP33 were recorded at the peak of the response on day 6 in lung ( Figure 1E , F and Figure S1A , B ) and spleen ( data not shown ) . CD8+ T cell numbers were reduced by a factor of 4 to 14 in old compared to young mice indicating that ageing also affected T cell immunity after VACV-GP infection ( B8R , A3L , A8R: p<0 . 01; GP33: p = ns ) . Compared to old naïve mice latent MCMV-infection in old mice further reduced VACV-GP induced CD8+ T cell numbers in the lung and spleen by a factor of 2 . 5 to 6 ( p = ns for all epitopes and organs; Figure 1E , F and Figure S1A , B for lung; spleen not shown ) . Overall , these data indicate that both ageing and latent MCMV-infection have a quantitatively comparable influence on antiviral T cell immunity in old mice . Both effects are modest but consistent and additive in old mice with latent MCMV-infection . Thymic involution is considered to be a major factor for T cell based immune senescence in humans [38] but in mice , thymic involution is considerably delayed in relation to their life expectancy [31] , [34] . Therefore , we next investigated whether Tx of mice at the age of 4–5 weeks would influence and enhance the development of immune senescence , particularly in the context of latent MCMV infection . We first tested whether MCMV control and MCMV-specific immune responses were comparable between Tx and non-Tx wild type ( wt ) C57BL/6 mice . Control of lytic MCMV replication was similar in Tx and wt mice ( Figure S2 ) and all mice were free of detectable virus in all organs tested and had thus established latent MCMV-infection within 6–8 weeks after MCMV-Δ157 infection ( not shown ) . We also compared the MCMV-specific CD8+ T cell responses longitudinally in wt and Tx mice , focussing on M45-specific CD8+ T cells , which usually display a conventional T cell kinetic of expansion , contraction followed by low level and stable memory T cell numbers , and M38-specific CD8+ T cells which display the phenomenon of memory inflation [20] , [22] , [24] . As shown in Figure 2A , C , memory inflation of M38-specific T cells occurred mainly during the first three months after infection and was not influenced by Tx . The apparent increase of M38-specific T cell frequencies in Tx-mice ( Figure 2C , G ) was caused by a loss of total CD8+ T cells after Tx and disappeared in the analysis of total T cell numbers ( Figure 2A , E ) . Interestingly , despite the abolished recruitment of new naïve T cells in Tx mice , M38-specific CD8+ T cells underwent a similar degree of memory inflation in Tx and wt mice ( Figure 2A , C ) . This confirms earlier findings , that thymic replenishment of the naïve T cell pool and the amount of recent thymic emigrants is not crucial for memory inflation [38] , [39] . Very advanced age was accompanied by an impressive loss of the inflated M38-specific CD8+ T cell population , both in normal and Tx mice ( Figure 2C , E , G ) . This could indicate that a large proportion of M38-specific memory T cells were reaching their proliferative limit ( Hayflick limit ) after >16 months of MCMV-infection at the age of about 20 months and that the naïve M38-specific precursors were exhausted as well . The conventional T cell kinetic of expansion , contraction and stable memory of the M45-specific CD8+ T cell population was not significantly influenced by Tx ( Figure 2B , D , F , H ) . M45-specific cells were stably maintained at low levels well into senescence with a slight trend for increasing numbers at very old age ( Figure 2D , F ) suggesting that these cells and their naïve precursors were not exhausted . To test and compare the impact of age , Tx and MCMV-infection on cellular and humoral immunity after a potent vaccination approach , mice were immunised with bacteriophage Qβ-derived virus-like particles ( VLP ) coupled to the LCMV peptide GP33 ( VLP-GP33 ) and adjuvanted with CpG-oligonucleotides [40] . Before immunisation , mice were thymectomised ( or not ) at the age of 4–5 weeks and 2–3 weeks later , they were infected ( or not ) with 107 pfu MCMV-Δ157 . Three or 15 months later , all mice were immunised with VLP-GP33 s . c . to induce a LCMV-GP33-specific CD8+ T cell response . On day 7 after VLP-immunisation , GP33-specific CD8+ T cell numbers were clearly reduced in the blood of old mice and of all Tx mice ( Figure 3A ) . Quantitatively , the effect of 12–15 months older age was equivalent to Tx at young age and the two effects seemed to be additive in old Tx mice which displayed the strongest reduction of GP33-specific T cell numbers after priming with VLP-GP33 ( Figure 3A ) and more clearly after secondary challenge with LCMV-WE ( see below , Figure 3B ) . Latent MCMV-infection did not have a measurable influence on VLP-induced T cell immunity , since GP33-specific CD8+ T cell numbers were already at the limit of reliable detection in old and in Tx mice ( Figure 3A ) . We next investigated , whether reduced priming efficacy after VLP-GP33 immunisation was associated with decreased T cell mediated antiviral protection after a challenge with LCMV-WE . On day 21 after VLP-GP33 immunisation , these eight groups of mice were infected intravenously with 2×103 pfu LCMV-WE . Eight days later , the GP33-specific CD8+ T cell response was analysed in several organs , including the lung ( Figure 3B ) . Despite reduced priming efficiency , the GP33-specific recall CD8+ T cell response after LCMV-challenge was comparable between aged and young mice in non-Tx animals . In addition , young Tx mice mounted similar T cell responses to non-Tx mice . Compared to all other groups , aged Tx animals had between 10- and 15-fold reduced numbers of GP33-specific T cells after LCMV-challenge . However , in none of the tested groups latent MCMV-infection had a measurable influence on the GP33-specific recall CD8+ T cell response . We also measured LCMV-titres in spleen , liver , lung and kidney of these mice . Consistent with the size of the CD8+ T cell responses , all non-Tx mice had nearly cleared LCMV from all organs tested including old and old/MCMV-infected mice ( Figure 3C for spleen , 3D for lung; filled symbols ) . When comparing LCMV titres after VLP-GP33 immunisation ( Figure 3C , D; filled symbols ) with LCMV titres without immunisation ( Figure 1A , B ) , it is apparent that VLP-GP33 immunisation was highly protective and was able to largely overcome the age- and MCMV-associated impairment of CD8+ T cell immunity . However , VLP-GP33 immunisation failed in old Tx mice , since LCMV titres were between 29- and 1290-fold increased compared to young Tx mice ( p = ns; Figure 3C , D; open symbols ) . In addition , old Tx mice with latent MCMV infection showed a trend to even higher LCMV titres in all tested organs ( 2- to 9-fold , p = ns ) and thus to reduced protective immunity compared to old Tx mice . Therefore , VLP immunisation was not sufficiently protective against a LCMV-challenge in old Tx mice with and without latent MCMV infection , despite a short interval after VLP priming and challenge with a low dose of a LCMV strain with intermediate virulence . To investigate T help dependent antibody responses in these eight groups of mice , the Qβ-specific IgG-response was measured 10 ( Figure 4E ) and 20 days ( Figure 4F ) after VLP-GP33 immunisation . In young mice , Qβ-specific IgG-titres were similar at both time points irrespective of Tx and/or latent MCMV-infection . In contrast , Tx and ageing clearly and cumulatively reduced Qβ-specific IgG-responses , particularly on day 10 after immunisation . Qβ-specific IgG-titres were between 5- and 15-fold lower in old/uninfected compared to young/uninfected mice , irrespective of Tx . On day 20 , the age-related differences in antibody responses were smaller indicating that ageing alone primarily delayed the response . In contrast , ageing combined with Tx resulted in a delay and an absolute reduction of the Qβ-specific IgG response . Latent MCMV-infection had a small but consistent impact on Qβ-specific IgG-titres in old mice early after immunisation ( day 10; Figure 3E , p = ns ) but not later . Therefore , latent MCMV-infection is rather delaying than suppressing the antibody response after VLP-immunisation . Overall , ageing and Tx clearly reduced the efficacy of VLP-immunisation whereas the influence of MCMV was very limited . We next investigated whether CD8+ T cell intrinsic or extrinsic factors were involved in impaired immune responses of aged , Tx and MCMV-infected mice . To this end , we co-transferred naïve TCR-transgenic ( tg ) CD8+ T cells specific for LCMV-GP33 and TCR-tg CD4+ T cells specific for LCMV-GP61 from young Ly5 . 1 positive mice into the eight different groups of Ly5 . 2 positive recipient mice mentioned before: young ( 4–6 months old ) or old ( 23–28 months old ) ; MCMV-infected ( for 2–4 months or 21–24 months ) or uninfected; Tx ( at 4–5 weeks of age ) or no Tx . One day later , all recipient mice were infected with 5×106 pfu VACV-GP i . p . Donor and recipient derived GP61-specific CD4+ and GP33-specific CD8+ T cell responses were quantified in the spleen ( data not shown ) and in the lung on day 6 after infection . Surprisingly , expansion of TCR-tg CD4+ and CD8+ T cells was comparable in all groups of recipient mice , although variability within groups was increased in old and in Tx mice ( Figure 4A , B ) . This indicates that MHC class I and class II antigen presentation pathways were sufficiently functional after VACV-GP-infection irrespective of ageing , Tx and latent MCMV-infection to allow comparable expansion of naïve TCR-tg CD4+ and CD8+ T cell populations . Interestingly , the endogenous GP33 and A8R-specific CD8+ T cell responses showed the familiar pattern: reduced expansion of virus specific CD8+ T cells was associated with age , Tx and latent MCMV-infection ( Figure 4C , D ) but only differences between young and old mice were statistically significant . Despite adoptive transfer of additional and functional LCMV-GP specific TCR-tg CD4+ T cells the endogenous GP33-specific CD8+ T cell response was not ‘rescued’ in old , old Tx or old MCMV-infected recipients . Since the GP33-specific CD8+ T cell response was shown to be T help dependent after VACV-GP-infection [41] , this suggests that lack of CD4 T cell help was not a major reason for poor CD8+ T cell expansion in these mice . Overall , our findings are best explained by an intrinsic defect of the endogenous CD8+ T cell population to expand after cognate antigen encounter due to ageing , Tx and MCMV-infection . To further investigate the mechanisms behind impaired T cell immunity in the context of ageing , thymic involution ( i . e . Tx ) and MCMV-infection we performed a detailed quantitative analysis of the impact of these three parameters on the composition of the peripheral T cell pools . There is compelling evidence from human studies that HCMV-infection leaves a unique signature on the CD8+ T cell compartment but less so for the CD4+ T cell compartment [26]–[28] . We first analysed the CD8+ T cell compartment in the blood of young ( 3–4 months ) , middle-aged ( 13–18 months ) and old mice ( 23–28 months ) with/without Tx and latent MCMV-infection , respectively ( Figure 5 ) . MCMV-infection of young mice resulted in a substantial expansion of the total CD8+ T cell compartment irrespectively of Tx and this difference was maintained during ageing ( Figure 5A ) . Independently of MCMV-infection or Tx , there was an age-associated decline of the total CD8+ T cell pool . Finally , Tx initially reduced the size of the total CD8+ T cell pool in young and young MCMV-infected mice , but this difference gradually disappeared over time . Several human studies have demonstrated that HCMV infection is associated with a sizeable reduction of the naïve CD8+ T cell pool , although most of these studies recorded T cell frequencies and not total numbers [26] , [28] . Indeed , our results confirm that MCMV-infection is associated with an immediate and pronounced reduction of naïve CD8+ T cell frequencies comparable to human studies ( Figure S3A ) . To our surprise , absolute numbers of naïve CD8+ T cells were similar in the blood of uninfected and MCMV-infected mice ( Figure 5B ) . These data were confirmed in another cohort of 8 and 13 months old mice , where absolute naïve CD8+ T cell numbers were not reduced by latent MCMV-infection in non-lymphoid ( lung , blood ) and in secondary lymphoid organs ( spleen , cervical and inguinal lymph nodes , Figure 6A ) . In contrast , ageing and Tx clearly and cumulatively reduced the total naïve CD8+ T cell pool leaving very few naïve T cells in old Tx mice . We suspect that the apparent increase of naïve T cells in old/Tx/MCMV-infected mice ( Figure 5B , right panel , right column ) might be caused by a poor discrimination of CD44-high and CD44-low T cell populations in this group of mice . In contrast to the naïve T cell pool , MCMV-infection induced a massive ( >10-fold ) expansion of blood CD44+ CD62L− effector memory ( Tem ) CD8+ T cells in young mice independent of Tx ( Figure 5C ) . During ageing , Tem-numbers slowly increased in uninfected and uninfected Tx mice and remained stable in MCMV-infected mice . Although the impact of MCMV-infection on Tem differences gradually dwindled with increasing age , Tem-numbers were consistently and significantly increased by MCMV-infection across all age groups , whereas Tx had no significant influence . Tem-numbers were also significantly increased in the lung and spleen of MCMV-infected mice whereas numbers were comparable to MCMV-naïve mice in both sets of lymph nodes ( Figure 6B ) . Contrary to Tem , the CD44+ CD62L+ central memory ( Tcm ) CD8+ T cell pool was not measurably altered by MCMV-infection in blood , lung , spleen or lymph nodes ( Figure 5D and 6C ) . In addition , neither ageing nor Tx had a measurable influence on absolute numbers of blood Tcm ( Figure 5D ) . In summary , ageing and Tx mainly reduced the naïve CD8+ T cell pool whereas MCMV-infection had surprisingly little influence on naïve T cell numbers but massively boosted the Tem pool . These additional T cells were accommodated in the CD8-compartment by a long lasting expansion of the total CD8+ T cell pool in MCMV-infected mice and not by a reduction of the naïve T cell pool . The total CD4+ T cell compartment was not significantly altered by MCMV-infection ( Figure 7A ) although , there was a trend to increased CD4+ T cell numbers in Tx mice with latent MCMV-infection . In contrast , both age and Tx significantly and cumulatively reduced the total CD4+ T cell pool . Similarly , latent MCMV-infection did not measurably influence the total number of naïve CD4+ T cells in the blood ( Figure 7B ) or in different lymphoid and non-lymphoid organs ( Figure 6D ) . Ageing was associated with a progressive but gradual loss of naïve T cell numbers whereas the very prominent Tx-associated reduction appeared within 6 months after Tx , leading to a >30-fold drop of naïve CD4+ T cell numbers from young to middle-aged mice ( Figure 7B ) . In the memory compartment , latent MCMV-infection also led to a slight expansion of blood CD4+ Tem but not of Tcm in young mice and this difference was partially maintained for Tem during ageing ( Figure 7C , D ) . However , in lung , spleen and lymph nodes CD4+ Tem and Tcm numbers were similar in MCMV-infected and MCMV-naïve mice ( Figure 6E , F ) . Ageing and Tx only had a minimal influence on the total number of CD4+ Tem ( Figure 7C ) . Ageing led to a gradual and progressive reduction of CD4+ Tcm in normal mice whereas this process was accelerated in Tx-mice leading to low Tcm numbers already in middle-aged mice ( Figure 7D ) . Overall , the effects of MCMV-infection on the CD4-compartment were much more subtle but similar to CD8 as an overall trend , whereas Tx had a more profound and immediate impact on the total and the naïve CD4 compartment .
Our data demonstrate for the first time in a mouse model that long-term latent MCMV-infection interferes with the induction of protective immunity in aged mice . Latent MCMV-infection additively impaired the poor control of LCMV-infection in old mice and this loss of antiviral protection was linked to reduced expansion of LCMV-specific CD8+ T cells . Moreover , latent MCMV-infection was associated with decreased CD8+ T cell expansions in old mice after infection with VACV but not after VLP-immunisation . Nevertheless , protective efficacy of VLP-immunisation against LCMV-challenge was impaired in old Tx mice with latent MCMV-infection . These data from a variety of infectious and non-infectious experimental models establish a causal role of latent MCMV-infection as a propagating factor for poor immunity in old age contributing to T cell based immune senescence . Our data are strongly corroborated by similar results from Cicin-Sain et al . , who found reduced CD8+ T cell responses after superinfection with Influenza virus , West Nile virus ( WNV ) and Herpes simplex virus Type 1 in old mice with latent MCMV-infection [42] . At first glance our findings seem to contradict an earlier study demonstrating partial protection against bacterial challenge in mice recently infected with MCMV or murine herpesvirus-68 by activated innate immune responses [43] . Although our study primarily focused on the impact of long-lasting latent MCMV-infection on T cell immunity in old mice , where we demonstrate impaired T cell mediated antiviral protection in latently infected mice , we have neither seen any protective effect of MCMV-infection against LCMV or VACV in young mice ( Figure 1A , B; VACV not shown ) . These discrepancies are most likely explained by differences in challenge systems: while Barton et al . used bacterial pathogens ( i . e . Listeria and Yersinia ) [43] we and Cicin-Sain have applied viral challenges [42] . Alternatively , partial protection by activated innate immune mechanisms may have been counteracted by impaired adaptive immunity in young mice with latent MCMV-infection . Our findings in this experimental mouse model strongly support previous observational human studies suggesting that HCMV-infection may be associated with accelerated immune senescence [12] . This notion was largely based on longitudinal cohort studies and cross-sectional epidemiological studies demonstrating a significant association of HCMV-seropositivity with decreased survival of very elderly [13]–[15] . Moreover , HCMV-infection was associated with reduced immunogenicity of influenza vaccination in elderly [17] although this was not confirmed more recently [16] . Interestingly , immunogenicity of CpG-adjuvanted VLPs was sufficient to largely overcome the age- and MCMV-associated immune attenuation in our experimental model ( Figure 3 ) . Only very old Tx mice with particularly low naïve T cell counts before VLP-immunisation were not protected against a LCMV-challenge ( Figure 3C , D ) This may indicate that more immunogenic vaccines than those currently available for influenza could help to improve protective immunity in the subset of elderly with a minimally maintained naïve T cell repertoire . All of the above mentioned human studies about the impact of naturally acquired HCMV-infection on waning immunity in the elderly are limited by the fact that a variety of factors may strongly influence any significant association . Potential confounders include 1 ) individual characteristics of HCMV-infection like virus isolate , dose , route , time point and severity of primary infection and the frequency of re-infection or re-activation , 2 ) genetic variability of the host directing the antiviral immune response , 3 ) general health and socio-economic situation of the host both presently and in the past ( co-morbidities , medication , nutrition , living conditions , physical activity , health seeking behaviour a . s . o . ) , and 4 ) the host's past infectious history apart from HCMV . For obvious reasons , it is impossible to control for most of these parameters in human studies . In contrast , we were able to exclude or balance these confounders in our mouse model . Therefore , our data provide the most direct evidence that CMV-infection itself is the driving force for the association between CMV-infection and impaired immunity in aged hosts . Concerning the major mechanism of CMV-enhanced immune senescence , most experts in the field favour the concept that CMV-driven memory T cell accumulations progressively restrict the size and the diversity of the naïve T cell pool , thus exacerbating the age associated alterations of the T cell compartment [44] . Studies in old mice and macaques have clearly established that both size and diversity of the naïve T cell pool are crucial for maintenance of immune protection [11] , [45] . Indeed , several human studies have shown that naïve CD8+ T cell frequencies in the blood were significantly reduced in HCMV-infected elderly or thymectomised young adults [26] , [28] , [46] but only few studies have reported significant reductions of absolute naïve CD8+ T cell numbers [27] . Our data demonstrate that total numbers of naïve CD4+ and CD8+ T cells were not significantly reduced by MCMV-infection in blood , spleen , lymph nodes and lung . This argues against the currently favoured hypothesis that CMV-infection reduces the number of available naïve T cells . At present , we cannot formally exclude the possibility of MCMV-dependent qualitative alterations of the naïve T cell pool , especially concerning repertoire diversity . Since adoptively transferred CD4+ and CD8+ T cells displayed comparable immune responses in all recipients irrespective of ageing , thymectomy and MCMV-infection , our results are best explained by a deficient activation and/or expansion of the endogenous T cell population ( Figure 4 ) . Indeed , Cicin-Sain et al were able to demonstrate an inhibitory effect of latent MCMV-infection on T cell recruitment and/or activation exclusively in draining lymph nodes after intranasal Influenza infection [42] . Our data further suggest that MHC class I and class II dependent antigen presentation was not substantially altered by ageing , thymectomy or latent MCMV-infection to interfere with activation and expansion of tg CD4+ and CD8+ T cells after VACV-infection . This is in line with similar antigen presentation in young and old monkeys after Vaccinia immunisation [45] . Although productive MCMV-infection has been shown to severely impair DC function [47] our results argue against a major interference of latent MCMV-infection with overall functionality of DCs as an explanation for reduced heterologous antiviral immunity: 1 ) Latent MCMV-infection of young mice was not associated with any measurable reduction of T cell immunity . 2 ) Activation , expansion and functionality of adoptively transferred naïve transgenic CD4+ and CD8+ T cells were comparable in all hosts tested , irrespective of MCMV-infection , age and Tx ( Figure 4A , B ) . 3 ) Latent MCMV-infection did not significantly impair the primary CD8+ T cell response after VLP-GP33 immunisation in old mice ( Figure 3A ) although protective efficacy was eventially reduced in MCMV-infected old Tx mice ( Figure 4C , D ) . Moreover , Cicin-Sain et al show that latent MCMV-infection with a mutant virus deficient for critical immune evasive genes was still associated with poor WNV-specific T cell immunity [42] . Lastly , it is unlikely that limiting T cell help was a major factor for MCMV-infection associated immune failure in our experimental setup . Provision of additional T cell help by adoptive transfer of tg CD4+ T cells was not sufficient to rescue the poor endogenous CD8+ T cell response in old , Tx and MCMV-infected mice although CD8+ T cell expansion after VACV-GP infection is known to be help dependent [41] . As opposed to the naïve T cell compartment , there is compelling evidence in mice , macaques and humans that CMV-infection drastically increases the memory T cell pool , particularly regarding CD8+ Tem [19] , [25]–[28] , [48]–[53] . Interestingly , those human studies indicating an association of HCMV-infection with poor vaccine immunogenicity or with reduced survival of very elderly , have also reported an association with CD8+ Tem expansions and not with a reduction of naïve T cells [15] , [17] . Similarly , the correlation of poor T cell expansion after heterologous WNV infection was stronger for increased Tem than for reduced naïve T cells in the MCMV-model [42] . Therefore , data from the MCMV-model support the human findings and demonstrate a vigorous expansion of Tem after MCMV-infection which was maintained into old age by memory inflation . MCMV-dependent Tem accumulation caused a significant and long lasting expansion of the total CD8+ T cell pool without restricting the available immunologic ‘space’ for naïve T cells explaining the discrepancy between reduced frequencies but maintained total numbers of naïve T cells after MCMV-infection ( Figure 5B , S3A ) . This highlights the fact that the CD8+ T cell compartment has considerable plasticity and can grow in size according to the cumulative antigen experience of the host [54] . Based on these findings , we postulate that CMV-dependent CD8+ Tem expansion is more important for CMV-enhanced immune senescence than naïve T cell suppression . Moreover , we would like to propose a model of enhanced competition for immunologic niches and survival factors between pre-existing CMV-specific memory T cells and newly generated effector T cells after heterologous infection or immunisation ( Figure 8 ) . CMV maintains memory T cells at very high frequencies well into old age , preferentially in the CD8+ T cell compartment as Tem , which persistently occupy critical niches both within secondary lymphoid ( i . e . spleen ) and peripheral organs . At first glance , the absence of Tem accumulations in lymph nodes of MCMV-infected mice ( Figure 6C , D ) may contradict a competitive model . However , Torti et al have recently shown that MCMV-specific inflationary T cells maintain a Tcm phenotype in lymph nodes where they interact with antigen presenting cells of non-hematopoietic origin and then egress the lymph nodes probably as Tem [55] . Therefore , competition of expanded MCMV-specific memory T cell populations with newly generated effector cells after heterologous infection or immunisation may still take place within , on the way out or even outside lymph nodes . Of note , we find the most robust differences in T cell immunity after systemic LCMV-infection ( Figure 1A–D ) , where T cell competition may preferentially occur in the spleen and not in lymph nodes . Moreover , LCMV-infection induces a massive inflammatory response which has the potential to trigger MCMV-reactivation . If this occurs expanded CMV-specific memory T cell populations are not only induced by bystander activation but also directly by TCR-mediated signals leading to further memory inflation and potentially to enhanced competition with heterologous T cell responses . However , our hypothesis of enhanced T cell competition needs to be tested in future studies together with a detailed analysis of the naïve T cell repertoire in old mice with and without latent MCMV-infection , since the absolute number of naïve precursor T cells is most likely to be CMV-independent but nevertheless crucial for maintained immunity in aged hosts . This is suggested by our findings of comparable T cell immunity in latently infected young mice , where the T cell repertoire was not yet affected by ageing or Tx , and supported by the comparable expansion of tg T cells irrespective of age , MCMV-infection or Tx , where we artificially increased and normalized the naïve precursor frequency . In summary , our results demonstrate that MCMV-infection has a moderate but significant negative influence on antiviral immune responses and protective immunity in old mice . However , Tx at young age and ageing itself seem to restrict immunity more profoundly than MCMV-infection . The involved mechanisms are likely to be different since ageing and Tx operate via the restriction of the naïve T cell pool while latent MCMV infection leaves the naïve T cell pool untouched but mainly propagates memory and total T cell expansions ( Figure 8 ) . Our model of latent MCMV-infection with or without Tx seems to be well suited to further investigate the underlying mechanisms and to develop and test preventive strategies . Finally , our results strongly support previous human studies that have found an association of CMV-infection with declining immunity in old age [12] . Since we and others were able to demonstrate a similar association in a very well controlled mouse model [42] there is now solid evidence that latent CMV-infection is a propagating factor for immune senescence in mice and in humans .
This study was carried out in strict accordance to the guidelines of the animal experimentation law ( SR 455 . 163; TVV ) of the Swiss Federal Government . The protocol was approved by Cantonal Veterinary Office of the canton of Zurich , Switzerland ( Permit number 174/2006 and 158/2010 ) . Surgery was performed under isoflurane anesthesia and all efforts were made to minimize suffering . C57BL/6 mice were purchased from Jackson Laboratory ( Germany ) and Charles River ( Germany ) and were kept under specific pathogen free ( SPF ) conditions throughout the study . Transgenic mice expressing the TCR specific for the LCMV-GP derived CD8+ T cell epitope GP33 ( P14 mice ) and mice expressing the TCR specific for the LCMV-GP derived CD4+ T cell epitope GP61 ( Smarta mice ) were provided by Prof . A . Oxenius ( ETH Zurich , Switzerland ) [56] , [57] . C57BL/6 mice were thymectomised at 4–5 weeks of age by vacuum extraction of the thymus along the trachea . Mice were anesthetized with isoflurane during the procedure and received analgesia with Buprenorphin i . p . from day -1 until day 6 after Tx . They entered the experiment after a recovery period of 2–3 weeks . All viruses were provided by Prof . A . Oxenius ( Zurich , Switzerland ) . The recombinant MCMV-Δ157 is a deletion mutant of MCMV lacking the m157 ORF leading to decreased NK-cell control in C57BL/6 mice [58] . The mutant virus was originally generated according to the published method using the bacterial artificial chromosome BAC pSM3fr [59] . MCMV was propagated on mouse embryonic fibroblasts ( MEFs ) and virus titres of virus stocks and organ homogenates were determined by plaque assays on MEFs as previously described , using centrifugal enhancement of infectivity and an adapted Avicel overlay ( 3% Avicel , 10% MEM , 10% FCS , 200 mM Glutamine , penicillin , streptomycin , 10 mM HEPES/NaHCO3 , H2O ) [60] . Mice were infected with 107 pfu of MCMV-Δ157 at the age of 6–10 weeks . LCMV-WE was propagated on L929 fibroblast cells . LCMV titres were determined by a virus focus forming assay on MC57G fibroblasts as previously described [61] . Mice were infected with 2×103 pfu LCMV-WE i . v . at the indicated time points . VACV-GP was propagated and plaqued on BSC40 cells . Mice were infected with 5×106 pfu VACV-GP i . p . at the indicated time points . The virus like particles VLP-GP33 were composed of the core protein of the Qβ-bacteriophage , coupled with the LCMV derived GP33-peptide , packaged with CpG oligonucleotides ( 5′-GGGGTCAACGTTGAGGGGGG-3′ , thioester stabilized ) and produced as previously described [40] . Mice were immunized with 200 µg VLP-GP33 s . c . into the flank at the indicated time points . The VLPs were provided by Dr . M . Bachmann ( Cytos Biotechnology , Schlieren , Switzerland ) . Adoptive transfer experiments were performed with cells isolated from spleens of tg Ly5 . 1-positive P14 or Smarta mice into Ly5 . 2 positive recipients to allow for simple discrimination of their origin by FACS . Using CD8 or CD4-beads , TCR-tg CD8+ or CD4+ T cells were purified by MACS from donor P14 mice or Smarta mice , respectively . 104 naïve tg CD8+ and CD4+ T cells were mixed and co-transferred i . v . into naïve or MCMV-infected recipient mice . One day later , mice were infected with 5×106 pfu VACV-GP i . p . and responses were analysed after six days . Infected or uninfected mice were anesthetized i . p . with an anaesthetic cocktail containing Xylazin ( Rompun ) , Ketamin ( Ketasol ) and Acepromazin ( Prequillan ) . After the surgical level of anaesthesia was reached mice were perfused with 5–10 ml of cold PBS to remove all contaminating blood from the organs . Organs were collected , cut into small pieces and digested by incubating for 20 min at 37°C with 3 ml of an enzyme cocktail ( x100 stock: 2 . 4 mg/ml collagenase and 0 . 2 mg/ml DNAse dissolved in RPMI medium with 10% FCS ) . After gentle mechanical tissue disruption by pulling the sample through an 18 G needle and additional 20 min incubation with 2 ml of fresh enzyme cocktail the sample was pushed through a cell strainer and centrifuged . After a Percoll gradient centrifugation to isolate viable lymphocytes the cells were washed twice , counted and resuspended in RPMI 10% FCS for further analysis . Single cell suspensions of spleen cells were produced by gently pressing the organ trough a grid of stainless steel with a plug . Viable cells were counted by trypan blue exclusion using Neubauer counting chambers . Monoclonal antibodies for flow cytometry assays were purchased from Becton Dickinson ( Switzerland ) or BioLegend ( Switzerland ) . The following antibodies were used: anti-CD8-Pacific Blue/PerCp ( clone 53-6 . 7 ) , anti-CD4-PE ( clone RM4–5 ) , anti-CD44-PE-Cy7 ( clone IM7 ) , anti-CD62L-FITC ( clone MEL-14 ) , anti-IFNγ-APC/PE-Cy7 ( clone XMG1 . 2 ) , and anti-CD45 . 1-PE ( Ly5 . 1 ) ( clone A20 ) . M38-/M45-specific ( MCMV ) and GP33-specific ( LCMV ) CD8+ T cells were detected by MHC class I tetramer staining using APC-conjugated tetramers . Tetramers were produced as previously published [62] . 50–100 µl of blood or 106 cells isolated from spleen or lung were stained with fluorochrome conjugated monoclonal antibodies for 20 min at 4°C . Tetramer staining was performed for 20 min at 4°C or 37°C . Using BD FACS Lysing Solution cells were fixed for 10 min at RT . After a washing step , conjugated antibodies were added to the cells for 20 min at 4°C and cells were washed and resuspended in 200 µl FACS buffer for analysis ( PBS , 5 mM EDTA , 2% FCS , 0 . 05% NaN3 ) . Samples were measured with a 6 or 8 colour BD FACS Canto II Flow Cytometer using FACS Diva software . The data files were analysed with Flowjo version 7 . 5 . 2 . Calibration with beads was used to measure and calculate total cell numbers . MCMV derived peptides M45985–993 ( HGIRNASFI , H2-Db ) and M38316–323 ( SSPPMFRV , H2-Kb ) , LCMV-derived peptide GP3333–41 ( KAVYNFATC , H2-Db ) , NP396396–404 ( FQPQNGQFI , H2-Db ) and Vaccinia derived peptides B8R20–27 ( TSYKFESV , H2-Kb ) , A3L270–277 ( KSYNYMLL , H2-Kb ) and A8R189–196 ( ITYRFYLI , H2-Kb ) were purchased from EMC Microcollections ( Tübingen , Germany ) . Cytokine production of activated cells was detected by intracellular cytokine staining . 2×106 tissue lymphocytes or splenocytes were stimulated with 10−6 M specific peptide for 5–6 h at 37°C in the presence of Monensin ( 2 µM ) or/and Brefeldin A ( 10 µg/ml ) . Cells were then washed with FACS buffer and stained for cell surface markers for 20 min at 4°C . Cells were fixed and permeabilized with BD FACS Lysing Solution ( 1∶5 dilution , 0 . 05% Tween20 ) for 10 min at RT and washed with FACS buffer . Cytokine specific antibodies were added to the cells for 20 min at 4°C . After washing cells were resuspended into 200 µl PBS with 1% paraformaldehyde and acquired with a FACS Canto II cytometer . Qβ-specific IgG antibodies were quantified by ELISA as previously described [63] . One-way ANOVA followed by Bonferroni post-analysis was used for group comparisons using Graph Pad Prism ( GraphPad Software , La Jolla , CA ) . The p-values are indicated in the graphs ( *<0 . 05 , ** p<0 . 01 , *** p<0 . 001 ) . | Cytomegalovirus ( CMV ) persistently infects 50–90% of the human population . After primary infection , constant immune surveillance is required to prevent CMV-related disease . During ageing , increasing T cell resources are expended to keep CMV under control . Recent human studies have suggested that this investment may come at the cost of accelerated immune senescence , a condition describing the age-associated decline of the immune system's functionality . In the present study , we have developed a mouse model to directly investigate whether and how CMV-infection might impair immunity of aged individuals . We demonstrate that old mice with long-lasting CMV-infection are more susceptible to viral infections than old mice without CMV since their virus specific T cell response is suppressed . Contrary to the prevailing hypothesis we found no indication for a CMV-associated shrinking of the naïve T cell compartment . Instead , CMV-infection precipitated a massive expansion of memory T cells . Thus , we propose an alternative mechanism of CMV-enhanced immune senescence based on T cell competition between CMV-specific memory T cells and de novo generated T cell responses . In summary , we provide the first direct evidence that CMV-infection is indeed a propagating factor for poor immunity in the elderly . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunizations",
"medicine",
"infectious",
"diseases",
"immune",
"cells",
"aging",
"and",
"immunity",
"immunity",
"t",
"cells",
"immunity",
"to",
"infections",
"immunology",
"biology",
"viral",
"diseases",
"cytomegalovirus",
"infection"
] | 2012 | Immune Senescence: Relative Contributions of Age and Cytomegalovirus Infection |
Spatial navigation requires the processing of complex , disparate and often ambiguous sensory data . The neurocomputations underpinning this vital ability remain poorly understood . Controversy remains as to whether multimodal sensory information must be combined into a unified representation , consistent with Tolman's “cognitive map” , or whether differential activation of independent navigation modules suffice to explain observed navigation behaviour . Here we demonstrate that key neural correlates of spatial navigation in darkness cannot be explained if the path integration system acted independently of boundary ( landmark ) information . In vivo recordings demonstrate that the rodent head direction ( HD ) system becomes unstable within three minutes without vision . In contrast , rodents maintain stable place fields and grid fields for over half an hour without vision . Using a simple HD error model , we show analytically that idiothetic path integration ( iPI ) alone cannot be used to maintain any stable place representation beyond two to three minutes . We then use a measure of place stability based on information theoretic principles to prove that featureless boundaries alone cannot be used to improve localization above chance level . Having shown that neither iPI nor boundaries alone are sufficient , we then address the question of whether their combination is sufficient and – we conjecture – necessary to maintain place stability for prolonged periods without vision . We addressed this question in simulations and robot experiments using a navigation model comprising of a particle filter and boundary map . The model replicates published experimental results on place field and grid field stability without vision , and makes testable predictions including place field splitting and grid field rescaling if the true arena geometry differs from the acquired boundary map . We discuss our findings in light of current theories of animal navigation and neuronal computation , and elaborate on their implications and significance for the design , analysis and interpretation of experiments .
In 1948 , Tolman employed two analogies to describe the prevailing classes of models used to explain the experimental data on maze navigation and learning obtained from rats [1] . Tolman likened the stimulus-response class of models to an old fashioned telephone exchange , where incoming calls are linked via connecting switches to outgoing messages . Stimulus-response connections which result in reward are strengthened . In contrast , Tolman was a proponent of the field theoretic or cognitive map class of models , in which the telephone switchboard was replaced by a “map control room” . Tolman asserted that sensory inputs “are usually worked over and elaborated in the central control room into a tentative , cognitive-like map of the environment” . The core issue seems to be whether animals ( including humans ) acquire and use a unified , multimodal spatial representation for navigation . Alternatively , can a model without a cognitive-like map of the environment explain animal navigation data ? One of the most ubiquitous navigation strategies in the animal kingdom is path integration ( PI ) , a process by which an animal uses an estimate of self-motion to update its location estimate [2]–[4] . PI works in principle under most environmental conditions . There is abundant theoretical and experimental evidence that PI requires stable allothetic directional information in combination with idiothetic motion cues [5]–[11] . Hence in general , PI is likely to be a multimodal process which combines a mix of information from vision , proprioception , vestibular or inertial organs , motor efference copy , and other sources depending on species . Therefore that the PI state is itself a multimodal representation . For example , it was recently shown in humans that PI output depends on a combination of visual and idiothetic motion cues in combination , not independently [12] . Clearly , experimental data is consistent with the multimodality property of a “cognitive map” . However , it is conceivable that a representation of the world is multimodal and yet modular , and hence fragmented . Recently , an insect-inspired model was proposed in which the navigation system consisted of independent modules [13] . During navigation , each active module produced a directional output , which fed into a recurrent neural network to output an overall heading direction . Behaviours such as shortcutting and landmark-guided homing were successfully explained using this model . Importantly , the authors argued that by maintaining modularity , this model did not require a “map control room” and hence did not resort to a “cognitive map” to explain a number of important navigation behaviours which have previously been used to argue in favor of a “cognitive map” [1] , [14]–[17] . Once acquired , a fragmented neural representation of the world seemed sufficient for effective navigation . The above model highlights the distinction between a “cognitive map” and “map information” i . e . , the ability to deduce position of an animal or landmarks from a neuronal ensemble code does not guarantee the existence of a “cognitive map” . It is possible to infer current position from either the PI module or landmark units of [13] , and even reconstruct an approximate map of the traversed environment . Hence “map information” was clearly present in the model , despite the absence of a unified “cognitive map” . An important aspect of the “cognitive map” debate is where an animal's neural system lies along a spectrum spanning complete modularity to full information fusion . In simplistic terms , one may consider neural systems and neural models as being less or more “map-like” according to the degree of information fusion . Focusing on this one aspect of the complex debate , it is clear that even the navigation modules of [13] implicitly assume information fusion from various sensory channels , and may be considered as having some map-like characteristics . On the other hand , there is clear anatomical and functional evidence for information segregation within both vertebrate and invertebrate brains , suggesting that full information fusion is neither necessary nor advantageous . In this paper , we focus on whether navigation modules as per [13] can , at least in principle , be used for effective navigation inside arenas with featureless boundary walls , in the absence of visual cues . Furthermore , we determine whether or to what extent fusion of iPI and boundary information may improve localization under these challenging conditions . Most behavioural and in vivo recording navigation experiments are conducted under light . Information from visual directional cues may lead to superior accuracy and precision in localization and navigation simply by allowing more accurate PI ( see later ) . Moreover , the advantage of using visual cues is inextricably confounded by the fact that other spatial information is also implicitly present in visual scenes . In both real and simulated arenas , rat-like navigation behaviours may be replicated without explicitly extracting any spatial layout information about the arena , simply by storing and comparing low resolution views [18] , [19] . Therefore , the presence of visual information may improve navigation performance in a number of interrelated ways , including PI performance . To circumvent this complication , we focus on a subset of experimental scenarios where visual information is absent or minimized . PI can provide an animal with a continuous location estimate , even when environmental cues are ambiguous or transiently absent . In practice , PI is subject to the accumulation of errors over time , whose error magnitude has been shown to be critically dependent on the computations used for updating the state of the PI system [11] , as well as the directional information which is used [8] , [9] . Some classes of PI models have been shown theoretically to be intolerant to noise [11] . In general , two necessary conditions for noise-tolerant PI are an allocentric reference frame ( world-centered ) , and static directional representations . An allocentric Cartesian PI system ( e . g . [20] ) is one example , where the axes are bound to world-centered directions . Importantly , there are at least two computational subclasses which satisfy both criteria , of which one is “ring-like” and one is “map-like” [11] . Therefore , a “cognitive-map” is sufficient but not necessary for accurate PI in an open field . Conversely , the need for accurate open field PI argues neither for or against the existence of a “cognitive-map” . In an open field , accurate PI using noisy idiothetic information ( termed iPI ) is impossible beyond a few steps [8] , [9] . In contrast , equally noisy compass information may be combined with idiothetic speed estimation for allothetic path integration ( aPI ) , preserving accuracy , and with significantly smaller positional variance [8] , [9] . Hence vestibular , proprioceptive and motor efferent signals are insufficient for open field PI , whereas vision , magnetoreception or other allothetic sensory channels are typically required . Hence even with a “map-like” PI system , the absence of visual or other compass information prevents accurate PI , raising the question of whether iPI can be used as an effective navigation strategy at all . PI and landmark navigation are complementary processes . Irrespective of the sensory information or neural circuitry , PI requires calibration by using cues in the environment to correct for errors built up during the PI process . Here , we call any set of cues which vary with location to be “landmarks” . Animals can use a wide variety of landmark cues ( e . g . visual , auditory , olfactory , tactile ) , which provide a mixture of positional and directional information for a given environment . Some landmarks are uniquely associated with a location or orientation in the world while others are less specific . According to [13] , the association of a PI state ( vector ) with each independent unique landmark results in an array of landmark units in memory which serves to guide navigation , without requiring a “cognitive map” . Boundaries may be considered a subclass of landmarks characterized by their geometric nature , but not associated with one specific point location . There is evidence that neural processing of boundaries may differ from other landmarks [21] , [22] . It remains an open question how a navigation algorithm could use boundary landmarks which restricts an animal's path , but which provides no other identifying information [23] . In the present work , we focus on the use of boundary information , not the process of its acquisition . Neurons which are preferentially active in particular positions or orientations in space provide a quantitative indicator of the stability of the animal's navigation system . In particular , if a neuron exists whose activity shows spatial selectivity that is stable over time , then it follows that computations required to maintain stable spatial selectivity must occur somewhere in the navigation system . In the rodent literature , at least four major functional classes of spatially-selective neurons have been identified . Hippocampal place cells [24] encode the rodent's location , cortical and subcortical head direction cells [25] , [26] encode the rodent's orientation , medial entorhinal grid cells [27] , [28] encode a multiplicity of regularly spaced rodent locations . There are also medial entorhinal border cells [29] and subicular boundary vector cells [30] , [31] which both encode the rat's relative location to barriers or boundaries . A subtype of the medial entorhinal grid cells encodes pose ( conjunctive location and orientation ) [28] , [32] , [33] . In the presence of visual information , the functional relationships between spatially-selective cell types are complex and intimately related to both task and available cues ( reviewed by [34]–[37] ) . A number of rat brain regions have been identified containing cells which represent head direction , and which form an interconnected head direction ( HD ) system [38] . The rate at which the HD tracking system degrades in darkness has been the subject of several studies [39]–[41] . Three important properties have been reported: 1 ) significant drift occurred after two minutes , 2 ) the angular deviation distribution was approximately zero-mean and symmetrical , and 3 ) the absolute angular deviation between consecutive two-minute sessions did not change significantly over time . These three observations suggest that the HD system drifts randomly and approximately at a uniform rate in the absence of vision . In contrast to the head direction tracking system , place and grid fields remain stable for half an hour or more during active exploration in a dark environment devoid of visual cues . Rat grid fields have been reported to remain stable in round arenas for up to thirty minutes in darkness [28] . Blind rats can generate and maintain stable place fields following exploration of stable landmarks placed within a round arena [42] . However , olfactory and tactile cues were not actively minimized in either study . In a follow up experiment to [42] , it was shown that even if odor cues were actively removed by cleaning of the arena floor , 10% of place fields remained stable , and about 50% remained , even over a period of 48 minutes in darkness [43] . Similarly , mice CA1 place fields were found to be stable in darkness in a 1 . 5 m diameter circular water maze , where floor odour cues were unlikely to be present . Place field stability was observed for two consecutive twelve-minute sessions [44] . Taken together , the above evidence suggest that vision is not essential for the rodent navigation system , for upwards of half an hour . Over short distances , iPI undoubtedly plays a role in navigation without vision [45] . However , given a head direction system which shows appreciable error ( drift ) beyond the first two minutes in darkness , can iPI explain place or grid field stability in the medium ( 5–10 minutes [46] , [47] ) to long term ( >30 minutes [28] , [42] , [43] ) ? Alternatively , can an independent landmark module , perhaps containing boundary information , be used to maintain stable place fields ? In fact , can any model assuming only iPI and boundary information explain place and grid field stability in darkness ? In summary , there is an active research field considering how PI interacts with environmental information . However , to date we are not aware of any studies which take a quantitative approach to studying the errors of iPI in arenas , the information provided by the arena geometry , or whether observed neuronal properties can be explained without fusing iPI and boundary information . We propose that a stable estimate of location can be maintained by animals for over half an hour without vision , by optimally combining idiothetic motion cues with a featureless boundary map – akin to Tolman's “cognitive map” , but contrary to [13] . We first model the accumulation of errors using only iPI . Using analytical derivations and simulations we show this iPI model cannot maintain place and grid field stability , assuming realistic neural tracking accuracies . Next , we present theoretical arguments showing that using arena boundary geometry without PI is insufficient for localization . Together these results show that any model which uses a modular or decentralized navigation system , including [13] , is incompatible with rodent neural data . Finally , we show using computer simulations and robot experiments how iPI in combination with boundary sensing and a geometric map enables long term stability of a location estimate in a number of arena shape configurations , demonstrating similarity to published experimental results . We demonstrate that the stability of simulated place fields depends on both the arena shape and size . A number of predictions are made regarding the behaviour of place and grid fields under environmental manipulations in darkness , which depend on the way in which iPI and boundary information is used . We discuss these results relating to known neuronal properties , implications on mammalian navigation models , as well as the design and interpretation of experiments .
Arena sizes and geometries were based on published experiments . In all simulations , circular arenas had a 76 cm inner diameter , corresponding to published experiments [40] , [42] , [43] , [46] . Unless otherwise specified , we used square arenas of the same area ( 67 . 4 cm width ) for comparison . Other rectangular arenas are individually specified . Since the arena walls were assumed to be homogeneous , the simulated rat was unable to identify which wall ( or wall segment ) it was close to . Therefore , wall contact information per se did not provide positional information beyond the fact that the simulated animal was somewhere along the boundary . Individual rat trajectories were described by a discrete time 2D correlated random walk model , with boundaries ( Fig . 1A ) . Simulated rats walked on average 5 . 4 m per minute . See Text S1 for trajectory simulation details , and Video S1 , Video S2 & Fig . S9 for an example of a simulated 48 minute trajectory . The errors in direction and speed estimates for the PI system were modelled as Gaussian random variables . The HD system was assumed to drift coherently but randomly , resulting in the PI system only having access to a single erroneous estimate of head rotation per step . From the results of [40] , and the trajectory model described above , the HD error standard deviation was estimated to be approximately per step or per second ( Text S2 ) . In the absence of direct experimental data , we assumed that linear step size estimation error was normally distributed with and independent of the angular displacement estimation error . Note that linear displacement estimation error makes a relatively small contribution to the overall positional uncertainty using iPI . For example , assuming straight line navigation in an open field using the error model described , linear errors account for approximately % of the asymptotic rate of positional variance increase ( substituting the error model parameters into the results of [8] ) . A particle filter model was used to approximate the Bayes-optimal combination of boundary and iPI information . A rat moving randomly in an enclosed arena will make contact with the boundary sporadically , in principle allowing it to localize to a region close to the boundary . Wall contact can also provide distance and orientation relative to the wall . In brief , the particle filter approximated the pose uncertainty distribution of the simulated rat during iPI through a population of pose estimates ( particle cloud ) . A particle cloud represented a finite sample from the true pose distribution . Each particle may be considered as one possible pose ( conjunctive position and heading ) , and its history may be considered as the simulation of one possible trajectory . During iPI , the stepwise increase in true pose uncertainty was modelled by randomly drawing values from the HD and step size estimation error distributions described earlier , and added to each particle's pose . Knowledge of the boundary limited the positional spread of the particle distribution , while boundary contact further reduced the unlikely particle pose estimates . Particles were weighted according to the likelihood that their pose explained current sensory ( or memory ) information , and then the particle population was redistributed according to particle weights . The resulting particle cloud provided a distributed estimate of current pose , having combined arena memory and arena contact information . See Text S3 , Fig . S3 and [48] for further details . The standard stochastic universal resampling procedure was used to update the particle cloud . In principle , this procedure produces a particle distribution which approaches the Bayes-optimal posterior distribution ( overviewed in Text S3 ) . Mathematically , this property is only guaranteed if the error models are available and correct . In simulations , these error models were assumed to be available to the rat's navigation system . Empirically , however , small deviations did not appear to cause large differences in the place stability index or simulated place and grid fields . For example , in the iRat experiments ( see later ) , neither the wheel odometric errors nor IR range sensor errors were precisely known . In the particle filter variant used for the iRat , the wheel odometric errors were overestimates , while the IR range sensor error magnitudes were not explicitly used . The particles were effectively ranked based on their relative consistency with sensory data , and a fixed fraction were culled during wall contact ( see Text S3 for further details ) . In those scenarios where the test arena differed from the training arena , a variant of stochastic resampling was also used for comparison with the standard form . The variant followed the standard method until the final step of assigning pose to the new particle cloud on boundary contact ( see Text S3 for further details ) . In this variant , only new heading was assigned , preserving the particle's original position estimate . When the test and training arenas were identical , this variant was inferior at localization compared to standard universal resampling . However , when the two arenas differed , this particle filter variant avoided large jumps in overall position estimates , and generated tessellating grid-like fields in a greater number of scenarios ( Results , Text S12 ) . To provide a performance metric for the particle filter navigation model which accounted for both accuracy and precision , we devised a simple intuitive index of position estimation stability , termed place stability . The mean squared distance of the particles to the true position is affected by the spread of the distribution ( precision ) and any systematic drift of the particle cloud ( accuracy ) . From information theoretic principles , the baseline is assumed to be a uniform distribution of particles throughout the arena ( maximum entropy ) . The place stability index at each time point is defined as where is the expected squared distance of the particles given a true current position , and is the expected squared distance between a uniform distribution of particles and . Using squared distances results in simple analytic solutions of and for circular and rectangular arenas ( Text S4 , Table S1 ) . A performance index of 1 implies a positional distribution equivalent to a Dirac delta function at the true location , while a uniformly distributed hypothesis of position results in an index of 0 . 5 ( chance ) . Indices below 0 . 5 may occur if the spread of the distribution exceeds the arena area , or if there is negative spatial correlation . The latter may occur , for example , if the spatial representation is rotated 180° about the center of the arena , relative to the true position . Since the simulated rat trajectories covered the whole arena homogeneously , it was possible to derive the expected place stability index given boundary contacts ( Text S4 , Table S2 ) . To understand how the particle cloud representation of place or the place stability index may relate to place fields , a simple model was used to simulate Poisson spike probabilities . The probability of a spike following each step was modelled as a Binomial process with where . The spike probability decreases monotonically from unity according to the distance r between the center of the particle cloud and ideal firing position . The center of the particle cloud was treated as the center of mass or Cartesian mean , i . e . , . This is also the position which minimizes the squared distance to all particles . In all simulations , cm , corresponding to the size of the pixel of analysis ( e . g . [43] ) . The size of was chosen to be sufficiently large to allow an adequate number of spikes to be generated during a simulated experiment , while being sufficiently small relative to the spatial resolution of the analysis procedure so that did not dominate the spatial spread of simulated spikes . Although it is somewhat arbitrary what constitutes an adequate spike count for analysis , we aimed to have approximately the same number of spikes as analysis pixels or higher ( 788 analysis pixels in the circular arena ) , in the majority of 8 minute periods and field locations studied . This was to avoid spuriously high spatial information values from low spike counts . For instance , one spike per pixel spread randomly across half of the analysis pixels yields a raw Skaggs spatial information content of approximately 1 bit/spike . The latter results from a low spike count rather than true spatial specificity . In addition to using the Cartesian mean , place field simulations were repeated using the polar mean for the circular arena simulations ( Fig . S5 ) . The polar coordinates of the particles were first averaged to give . Here , the angular mean where denotes the 4-quadrant arctangent . The Cartesian coordinates of was used as a substitute for . The polar mean was used due to the fact that the particle cloud distribution in circular arenas tended to follow a crescent shape approximately aligned with the circular boundary ( discussed further in Text S9 , see Video S1 & S2 for an example ) . Under these conditions , the polar mean was a good approximation of the modal position of the distribution . The Cartesian mean was often close to or even within the concavity of the crescent-shaped distribution , rather than near the mode . This caused an underestimation of the radial position of the cloud . However , the cloud distribution tended to be a convex shape in rectangular arenas , so there was little difference between the two methods near rectangular boundaries . The polar mean was not used throughout the simulations because the estimation of radial distance close to the arena center was contaminated by the spread of the cloud ( Text S9 ) . Over a period of time , the simulated spike pattern represents a temporal average reflecting a sequence of complex particle cloud states . These states in turn depended nonlinearly on the actual trajectory taken , the boundary information gained , and random errors . Therefore , place stability changed dynamically during each trial , and affected the spatial specificity of the simulated spike sequence depending on location . Both positional and angular specificity were quantified using Skaggs information [49] to be comparable to published data on place fields . A maximum likelihood factorial model [50] was applied to check whether decoupling of positional and directional information affected the estimated spatial information content . Since our results showed that the particle filter output had a complex dependence on the pose distribution and wall information , we investigated the effect of using an arena boundary representation different to the one being traversed . This is analogous to a change in arena size and/or aspect ratio while in darkness . With vision , rat grid fields have been reported to rescale when rats are transferred between rectangular arenas of different size and aspect ratios [51] . Grid fields were modelled as multiple independent place fields distributed as a regular hexagonal tessellating array over the entire training arena . Grid fields were simulated by assuming that the firing probability was determined by where was the position of mode j of the grid field . Similar to place fields , the firing probability of each contributing subfield was given by . Following each step , the maximum allowable number of spikes was capped at one . It was assumed the training arena's boundary representation remained in memory during all tests . To show that the derived and computer simulation results can be applied in real environments , we used the prototype iRat [52] , [53] , Intelligent Rat Animat Technology robot for experiments in real arenas ( Fig . 2A ) . The iRat is comparable in size and mass to a laboratory rat at 150 mm×80 mm×70 mm at 0 . 56 kg . The iRat has a camera , speakers and microphone , on board computation via 1 GHz PC , WLAN , and IR ( infrared ) proximity sensors . The IR sensors may be considered as providing crude ‘whisker’ information near walls . In this study , only the three IR sensors were used to obtain three distance estimates when close to arena walls ( Fig . 2B ) , whereas the camera was not used . See Text S5 for details of the iRat experiments , and Text S3 for details on the particle filter variant .
Using the simplest description of locomotion which consists of a turn and step , it has been shown previously that the asymptotic rate of increase in positional variance per step is where is the mean step length , is the variance of the step length , and is the angular error per step [8] . This result was derived assuming iPI along a straight course in an open field . For a zero-mean , normally distributed , where is the variance of the HD angular error per step . For ease of interpretation , the variance rates in this section are reported in terms of time rather than steps . Let be the mean distance travelled per second . Since is the area of the traversable region within one second , iPI errors are considered irrecoverable if the positional variance increased beyond this rate . This is because the true position may be anywhere within an area too large to be traversed even in theory . Note that represents a highly optimistic threshold since one unit of positional variance encompass less than half of all possible positions in a circular bivariate Gaussian distribution ( Fig . 3A red dotted line ) . For a 95% confidence region , the threshold is where ( Fig . 3A red dashed line ) . Even with this correction , the threshold is optimistic since it represents the limit of search recovery ( assuming error-free search ) and could not plausibly sustain a stable place representation . Nevertheless , it allows estimation of a loose upper bound on the time limit of the use of iPI . Combining the results of [8] with the HD model in our current simulations , we determined whether it is theoretically plausible for iPI to maintain an accurate long term estimate of position . Assuming the HD cell error model described in Methods , without any step length estimation error ( ) , the predicted asymptotic positional variance increased at per second ( Fig . 3A black dashed line ) . Since , the positional uncertainty increased much faster than the maximum area which can be traversed , clearly showing that iPI cannot be used to accurately track movement along a straight trajectory in the long term . In the short to medium term , positional variances of iPI increases more slowly than the asymptotic rate [8] , [9] . Substituting the HD error model parameters into the exact variance expressions derived in [8] , the optimistic limit of was exceeded after 51 seconds ( perpendicular to axis of intended locomotion ) , and 192 seconds ( along the axis of intended locomotion ) , again demonstrating that iPI became irrecoverably inaccurate well within 1 to 3 minutes from the start ( Fig . 3A black solid lines ) . Next we considered tortuous trajectories where the intended path had directional variance ( Methods and Text S1 ) . Clearly , a rat trained to forage within a small arena has to change direction regularly , following a tortuous rather than straight course . This occurs , for example , in many experiments within confined arenas . Path tortuosity may decrease the rate of positional variance increase in two ways . Firstly , if an iPI system is unable to track the actual turns , then the HD error angle would be dominated by the physical turn angle i . e . , . For the path tortuosity used in the simulations in this work , the predicted asymptotic positional variance increase was ( Fig . 3A green dashed line ) , still greater than the conservative limit of . The limit of was exceeded within 10 seconds from the start of iPI ( Fig . 3A green solid lines ) . Another possibility is that path structure itself influences the way in which a small cumulative HD error impacts on the position estimate . This was modelled as an unbounded correlated random walk [6] with path directional standard deviation and HD error . Since closed form solutions to these variance functions are not available ( but see [6] for empirical approximations ) , Monte Carlo simulations were performed . The positional variance rate remained within the limit of for nearly eight minutes ( Fig . 3A blue lines ) . However , the 95% confidence interval exceeded the limit of in 88 seconds , making accurate iPI impossible within the first one and a half minutes . Nevertheless , an iPI system with small HD error can track a tortuous path more accurately than a straight path , showing that path structure itself can affect navigation performance . Inside an arena , animal paths are further constrained by the boundary . Since boundaries force animals to make turns that they would otherwise not need to in an open field , we expected that positional variance would increase more slowly inside bounded arenas , for a given baseline tortuosity of the intended trajectory . We tested this prediction using the trajectory model described in Methods and Text S1 . The maximum average rate of position variance increase was found to be less than . At first , this may seem to support iPI as a plausible process to maintain place stability inside bounded arenas . On more careful analysis , this was found not to be the case . Assuming a Gaussian distribution , we found that the 95% confidence area of the positional error distribution exceeded the entire area of the arena by 175 seconds . To quantitatively track the accuracy and precision of the position estimate of individual trajectories , we used the place stability index ( Methods , Text S4 ) . In a circular arena , the average over 103 simulations fell below chance ( 0 . 5 ) within 179 seconds ( Fig . 3B ) . Similar results were obtained for the square arena ( 184 seconds - see Fig . S6B ) . This means that despite a much reduced rate of increase of positional variance , the best estimate of position afforded by iPI is no better than chance level beyond 3 minutes . In other words , the navigation system has no useful information about its true location within the arena beyond 3 minutes without vision . Taking a third approach , we quantified the maximum spatial information of the position estimate itself , assuming a Gaussian field centered in a circular arena , and the trajectory and HD models described earlier . It was found that the corresponding spatial information fell below 1 bit after 150 seconds using iPI alone ( Text S6 , Fig . S1 ) . The maximum spatial information content of any firing field will be expected to fall below 1 bit per spike in under 3 minutes without vision . In comparison , some place fields in the “dark+cleaning” condition of [42] showed spatial information of about 1 bit/spike or higher over three consecutive 16 minute windows . In summary , iPI alone cannot sustain a useful position estimate beyond 2 to 3 minutes without vision in either open field or enclosed arenas . Intended path tortuosity and arenas can both reduce the magnitude of positional uncertainty , with implications for the comparison of results obtained in confined arenas and open fields . Interestingly , even aPI was unable to maintain place stability or generate a stable place field if used alone ( Text S11 , Fig . S7 , Table S4 ) . Assuming the same HD error distribution as above , but reset following each step to mimic having stable distant visual landmarks , the time taken for the average place stability index to drop below chance level ( 0 . 5 ) was increased but still unable to explain stable place or grid fields beyond approximately 5 minutes . The rat's navigation system was assumed to have continual access to information about whether it was in contact with the arena boundary or not . We investigated whether this information alone can increase the place stability index above chance level , assuming error-free knowledge of the arena size and shape . When wall contact has occurred , denoted by W+ , the ideal posterior positional distribution is a narrow region along the perimeter of the arena , since positions closer to the center of the arena should not result in wall contact . We investigated the possibility that this information alone may have increased place stability above chance by finding when wall contact has occurred . The range and expected IP values can be found assuming a uniform sampling of the perimeter , and assuming an idealized uniform posterior distribution following wall contact , corresponding to the perimeter line ( see Text S4 and S7 for further details ) . Under the above assumptions , in any square arena , while the expected or average . For any circular arena , . Note that these indices are independent of arena size . When not in contact with the boundary , denoted W− , the animal may be anywhere within the arena giving . Thus , in the absence of PI information , the expected place stability index does not exceed chance level ( 0 . 5 ) even assuming ideal information about the arena boundary ( summarized in Table S2 ) . Hence above-chance place stability cannot be achieved only using arena boundary information . An alternative interpretation of the above results is as follows . Suppose that a randomly foraging rat occasionally contacts the arena boundary . Mostly , the rat knows it is not at the boundary , so its internal representation is a uniform distribution over the entire arena , called the null position estimate . On boundary contact , its internal representation of position becomes a uniform distribution along the featureless boundary , called the boundary position estimate . We ask whether the boundary position estimate improves the estimate of current position when the rat is actually at the boundary . It can be shown that for all convex arena boundaries , the estimation error is actually greater if the animal used the boundary estimate compared to the null estimate ( see Text S7 & Fig . S2 for proof ) . This conclusion applies to all functions of position estimation error which increase monotonically with the estimation error distance . As an example , the mean squared position estimation error of an animal at the boundary of a circular arena using the boundary estimate is 33% larger than the null estimate ( Table S1 ) . Therefore , using the boundary geometry actually increases the mean squared position estimation error compared with using the null estimate . Inside arenas , positional uncertainty may be modified by the use of boundary information . Two types of information are considered: a ) boundary memory only; and b ) boundary memory plus wall contact information . The latter may be due to whiskers or other haptic information and was assumed to provide approximate wall distance and incident angle ( Methods ) . In both the circular ( Fig . 3B–D ) and square arena ( Fig . 3D , S6B & S6D ) , the average place stability index remained above chance for 48 minutes when arena boundary information was used . On average , using wall contact information improved place stability . In contrast , place stability dropped below chance within the first 8 minute window , in the absence of arena boundary information ( Fig . 3B , S6B blue lines ) . Boundary information in the square arena consistently improved average place stability beyond that in circular arenas . This pattern of results persisted when only the most stable 10% of trials in the two arenas were considered ( Fig . 3D ) . The average place stability indices remained above chance level for 48 minutes without vision in both circular and square arenas . Since it was shown earlier that neither iPI nor boundary information alone could achieve this , the current particle filter implementation demonstrates greater place stability than independent iPI and boundary landmark modules . For completeness , we consider the possibility that an animal's navigation system can switch between functional modularity and information fusion . We therefore ask whether an occasional switch to using a modular navigation system as per [13] may still produce a similar level of place stability , provided that optimal information fusion occurs at all other times . It can be shown that following a single boundary contact using a modular navigation system , the maximum expected place stability ( proof in Text S8 ) . Although this value is marginally higher than chance ( 0 . 5 ) , it is significantly lower than near-optimal fusion of iPI and boundary information after 48 minutes without vision ( t999 = 11 . 17 , p<10−100 ) . Most importantly , the uncertainty distribution becomes a circular annulus concentric with the boundary , incompatible with any stable place field restricted to one sector of the arena . This mechanism can , however , maintain a stable place field at the center of the arena . Overall , even occasional use of a modular navigation system causes a significant decline in place stability , incompatible with stable place fields ( other than at the arena center ) . Despite the higher place stability index with fused iPI and boundary information , the average place stability decreased continually over 48 minutes without vision in the circular arena . This decrease was clear even when the most stable 10% of trials were considered ( Fig . 3D ) . We investigated whether the decreasing place stability could support stable firing fields , using a Poisson probability model ( see Methods for details ) . Fig . 4A shows the pooled average firing field generated during consecutive 8 minute time windows , for the 10% of trials with the highest place stability indices ( to be comparable to the results of [43] ) . A more extensive set of simulated place field locations are shown in Fig . S4 & S5 corresponding to using the Cartesian and polar mean , respectively , to estimate particle cloud position ( see Methods and Text S9 for details ) . During each 8-minute time window , the firing fields were quantified using five metrics: 1 ) the spatial information content; 2 ) the directional information content; 3 ) the number of spikes; 4 ) the spatial correlation coefficient calculated bin by bin , relative to the first 8-minute window; and 5 ) the spatial coherence [54] ( Table 1 ) . The information content was calculated using the maximum likelihood factorial model [50] . The bin sizes used were 2 . 5 cm by 2 . 5 cm for position and 6° for direction . With boundary memory and wall contact information , the spatial information content remained close to 2 bits per spike for 48 minutes without vision , while the directional information content remained at less than 0 . 1 bit per spike . There was also high field coherence and the spatial correlation remained above 0 . 5 . Using the polar estimate of position , the spatial correlation remained above 0 . 8 for all fields and time periods except those at 30 cm ( Table S3 ) . The high spatial correlation is comparable to rat place fields in circular arenas of the same diameter , in the presence of visual information ( R = 0 . 70 [55] ) . Together , the simulation results show that in the circular arena , stable place fields are maintained for 48 minutes without vision , in at least 10% of trials , similar to experiment [43] . Place fields were considered stable by [43] as those which rotated by less than 12° between the control period and the first test period . This was estimated by rotating fields from each time window , about the arena center , to find the maximum spatial correlation Rmax possible . This angular displacement , Δθmax , is indicative of one possible way in which place fields may become unstable . Using the same analysis method but at 1° rather than 6° resolution , and using 8-minute instead of 16-minute time windows , we found that the most stable 10% of place fields rotated by 12° or less between the first period of no vision , compared with each of the subsequent periods , compatible with experiment ( see also Table S3 ) . These results further support the current model as a reasonable approximation of the computations carried out by the rat navigation system . For completeness , we tested whether boundary contact per se was beneficial ( Fig . 3B , Fig . 4B , Table 1 ) . The same procedures were used , but without boundary contact information . In the absence of boundary contact information , the spatial information content was substantially and consistently lower than with boundary contact ( see Fig . S6 and Text S10 for square arenas ) . Similarly , the spatial correlation was below that of having boundary contact information , for all time windows . In particular , the spatial correlation for the 8–16 minutes without boundary contact information was below that of the 40–48 minute time window with boundary contact information showing a significant , immediate and persistent decline in place field correlation compared to the first 8-minute period . The spatial firing pattern was less stable without arena boundary contact information . Hence boundary memory was useful in culling particles which went outside the boundary extent , thereby limiting the growth of the uncertainty distribution ( see also Text S3 ) . This contrasts from pure iPI , where the particle cloud width increased without limit . However , particles within the arena but far from the boundary were never culled in the absence of boundary contact information , even if their pose estimates were otherwise highly inconsistent with the sensory information from boundary contact . Thus boundary contact per se may be considered as providing information which , when used appropriately , can reduce positional uncertainty beyond that of having arena memory only . Consistent with analyses presented earlier , simulations using iPI alone resulted in no stable place fields ( Fig . 4C ) . Due to extremely low spike counts even with pooling across trials , the spatial information content could not be estimated reliably . A common experimental manipulation for testing neural and behavioural properties in navigation tasks is to change the arena size and shape . In simulation , this was achieved by explicitly specifying a different arena size and/or shape to that which was traversed . In this way , the arena in memory may be considered as that acquired during training , while the test arena is introduced at the beginning of each trial , at the moment when visual information becomes unavailable . It has been shown that some place fields established in a square arena either stretched or split when the arena geometry was changed [56] . More detailed analysis of the split fields showed that the firing subfields had different modal positions depending on the direction of travel of the rat . Although these experimental results were obtained with vision , we tested the effect of the same arena geometry manipulations without vision as predicted by the two particle filter models described in Methods ( Fig . 5 ) . Using the same place field model as described in Methods , we tested the effect of having a different traversable arena to that stored in memory . Directional information content was low in all cases ( Table 2 ) . Place field stretching or splitting was found in the three novel test arenas , with the emergence of directional selectivity in the split fields similar to [56] . In our simulations , the spatial information content decreased by more than 2 bits/spike between the training arena ( 61 cm by 61 cm , Fig . 5A & 5C upper left panels ) and horizontal rectangular arena ( 61 cm by 122 cm , Fig . 5A & 5C upper right panels ) , without a concomitant change in directional information . Hence the directional selectivity of the individual modes of the bimodal firing field ( Fig . 5B & 5D ) cannot be attributed to a change in the overall directional selectivity of the field . To determine whether the most recent wall contact may be related to the pattern of firing , the frequency of each immediately preceding wall contact was found ( Fig . 5B & 5D ) . In both fields , the highest frequency of recent contact was of the top wall , which was also the nearest wall . The two particle filter variants used yielded similar results . The largest relative differences in frequencies were of recent contacts with the left and right walls with over threefold changes consistently . In particular , during rightward traversals ( Fig . 5B & 5D upper panels ) spikes were preceded most recently by left wall contacts more frequently than right , while during leftward traversals , ( Fig . 5B & 5D lower panels ) spikes were preceded most recently by right wall contacts . These results show that the proximity of boundaries is correlated with the relative frequency of most recent contact , determined retrospectively from each spike . The marked differences in the relative frequencies when fields are divided based on rightward versus leftward trajectories can be explained as follows . Due to the temporal correlation of headings along simulated trajectories , a leftward trajectory is more likely to have recently come from the right part of the arena , and vice versa . Therefore , a leftward trajectory was more likely to be preceded most recently by contact with the right wall than the left wall , and vice versa . In the results shown in Fig . 5 , the arena representation in memory was a 61 cm by 61 cm square , and the place field was positioned 15 . 5 cm from the left wall and 45 . 5 cm from the right wall . Therefore during testing in the 61 cm by 122 cm arena , rightward trajectories resulted in maximal firing approximately 15 . 5 cm from the left wall ( Fig . 5B & 5D upper panels ) , while leftward trajectories resulted in maximal firing approximately 45 . 5 cm from the right wall ( Fig . 5B & 5D lower panels ) . When the test arena was a 122 cm by 122 cm square ( Fig . 5A & 5B lower right panels ) , the split fields showed particularly low spatial information content ( <0 . 3 bits/spike ) using either particle filter variant . The low spatial specificity was due to the large discrepancy between the dimensions of the training and test arenas in both spatial dimensions ( see Text S12 & Fig . S8 for further details on error mechanisms ) . With vision , grid field spacing has previously been shown to partially rescale along the direction of a rectangular arena which is stretched or compressed varying with the arena geometry transformation [43] . The rescaling factor in grid spacing was consistently less than that of the arena rescaling . We tested the same arena transformations using the particle filter model variants described in Methods . Firstly , we found that an unstable HD system ( e . g . without vision ) can maintain a variety of stable grid-like firing fields , even if the test arena differed to the training arena in geometry ( Fig . 6 & 7 ) . Secondly , using particle resetting of heading only , arena compression caused a partial rescaling of grid spacing ( Fig . 7A & 7B ) , in a manner qualitatively similar to that observed in vivo , with vision . The magnitude of the partial rescaling was less than reported ( approximately 25% of the arena rescaling , compared to 48% reported by [51] ) . Grid rescaling did not occur in simulations where arenas were stretched ( Fig . 6C , 6D , 7C , 7D ) . Instead , grid field splitting was seen - analogous to the phenomenon of place field splitting reported earlier . It must be emphasized that the primary purpose of simulating arena manipulations was to test whether it is possible for stable place and grid fields to be maintained without vision , despite different dimensions between the training and test arena . A secondary goal of these simulations was to demonstrate that specific hypotheses about the combination of iPI and boundary information can be modelled using the particle filter approach . The differences in results between the two variants of the particle filter used highlights the importance of determining the precise manner in which information is used for spatial navigation . To demonstrate near-optimal navigation without vision in real world conditions , we adapted the particle filter method to a mobile robot platform , the iRat , moving in a real arena ( Fig . 8 ) . Cumulative odometric errors caused a gradual drift in the estimate of heading and position , in an analogous way to simulated rodent iPI described earlier , making localization for any prolonged period of time using pure iPI impossible . In contrast , application of the particle filter to combine arena geometry and IR ‘whisker’ information maintained highly accurate localization for the duration of the experiment ( 5 minutes ) . All iRat experiments shown were conducted without using the on-board camera . There was no appreciable decrease in place stability , or increase in pose error , over the trial period . Successful localization using the iRat in a real arena demonstrates that the proposed algorithm is robust and can function without precise knowledge of real world characteristics including iPI estimation errors or wall contact estimation errors ( discussed in Text S3 & S5 ) . Egomotion and sensory error characteristics are often difficult to ascertain explicitly by a navigating agent , and may depend on interactions with the environment during a particular navigation trajectory which cannot be known ahead of time . This work also demonstrates the feasibility of using the iRat to test computational algorithms in similar arenas to rodent experiments .
It is possible that differences in experimental conditions account for some of the discrepancy between HD instability and place/grid field stability . The clearest test of this possibility should involve simultaneous recordings of HD cells , place cells and grid cells without vision , but has not been reported to date . It is possible that odor or other non-visual cues may provide orientation information . Ideally , experiments should be performed with stringent control and quantification of environmental cue signals . But like simultaneous in vivo recordings in multiple regions , such experiments may be challenging in practice . It is worth noting that the HD error rate was estimated from experimental data obtained without the active removal of odor cues [40] . Given that place field stability is adversely affected by odor cue removal [43] , it is possible that HD system stability may also be affected , potentially making the HD error rate larger than in our model . On the other hand , tuning of individual cells in the HD system may only be partially correlated , especially across multiple brain regions . If the observed HD error per cell has a random component which varies independently between HD cells , it is possible that the directional information error from the entire HD ensemble may be smaller than predicted based on individual HD cells . Based on published experimental evidence [26] , [56] , this seems unlikely to be of significance . If stable place and/or grid fields without vision imply the presence of a near-optimal distributed pose estimate , and this pose estimate is available to the HD network , then in principle , feedback may correct HD cell tuning errors . This seems unlikely in practice . Firstly , there does not seem to be any anatomical or functional evidence reported of feedback correction to the HD cells from either place cells or grid cells . Secondly , our simulations predict that the optimal heading error is small and relatively stable in square arenas ( Fig . S6F , S6H ) , leading to the prediction that a fully corrected HD system should not drift in square arenas without vision . Since the HD system apparently drifts even in a radial arm maze [29] , this prediction seems unlikely but remains to be tested . Thirdly , in open field PI with vision or other compass cues , the absolute HD ( even if with noise ) is more accurate than angular displacement estimates ( e . g . , refs [8] , [9] ) . This is especially relevant if familiar landmarks are sparse . Hence there is a role for absolute HD , as well as AHV information in a spatial navigation system . An effective head direction system should contain both . Finally , it has been demonstrated previously that conjunctive grid cells have the computational properties needed to represent pose [33] . Since conjunctive grid cells are found in mEC [28] , [32] , it seems plausible for the pose estimate to be maintained in mEC , and sustain place field stability downstream . It is worth noting that during PI , the correct update of pose is in the direction of translation which may not necessarily be the same as head direction . However , neurons which encode translation direction per se have not been reported . It is also unclear how grid and/or place cells may switch between the types of direction information to use . Although an important question , it is outside the scope of our current modelling efforts . We speculate that removal of visual input may trigger a switch in , or at least modulate , the type of directional information used . It seems plausible that some olfactory information was present during the place field recordings in the “dark+cleaning” experiments reported by [43] . Despite the arena being cleaned prior to switching off room lights , it was possible that the rats laid down odor cues strategically or accidentally during the 48 minutes in darkness . However , HD cell tuning curves continue to drift under similar conditions [40] , [41] suggesting that natural odor cues do not provide absolute orientation information . Furthermore , blind rats do not express stable place fields in circular arenas until at least one haptic landmark is introduced within the arena [42] suggesting that natural odor cues are insufficient to generate stable place fields . More recently , place fields in mice were found to remain stable in darkness in a Morris water maze [44] , suggesting that stable ground odor cues are not necessary to maintain stable place fields in darkness . Nonetheless , once formed , the maintenance of stable place fields may be aided by the presence of olfactory cues in the absence of vision [43] . Similarly , olfactory or other allocentric cues may have contributed to the stability of grid fields in darkness reported by [28] . Therefore it is important to note that our work did not assume odor cues to model stable place and grid fields in darkness . If the head direction system is truly unstable , the HD tuning direction per se cannot be used to maintain a stable place representation via PI . This is because angular displacement errors accumulate , which is incompatible with medium to long term localization ( Figs . 3 , 4 , 1 , S6 , S7; see also [8] , [9] ) . Angular displacement or angular velocity , however , can be used since a probabilistic estimate of pose can be updated without using absolute direction . From moment to moment , this type of navigation may still be considered as an iPI process but with important and frequent corrections using a combination of boundary memory and boundary contact information . The latter provide a mix of conjunctive position and direction information which can be used to improve the distributed estimate of pose in an approximately Bayes-optimal way . It is an approximation because a particle filter , by its nature , is necessarily a discrete approximation of pose even if all filter properties are optimized for known sensory error characteristics . Computationally , angular displacement or angular velocity estimates may be obtained via differences between consecutive HD readings , or more directly via signals of angular velocity as found in angular head velocity ( AHV ) cells [38] , [57] , [58] . In this work , we have assumed that the rate of drift of the entire HD system , including AHV cells , is indicated by the drift in HD tuning functions . It may be the case that AHV signals have a different error rate compared to the difference between HD tuning signals . However , in the absence of vision and other compass cues , it seems highly likely that the HD signal itself is maintained indirectly through an estimate of angular velocity or acceleration . By providing angular displacement , an erroneous HD system is computationally compatible with a stable place or grid representation . As shown , stable place and grid fields can be simulated using a distributed conjunctive representation of position and orientation ( pose ) , when combined with a coherent representation of the arena boundary . The former is consistent with recent neurorobotic research suggesting that a distributed conjunctive pose representation is required for managing perceptual ambiguity using visual sensing [33] . Here , we have shown that managing perceptual ambiguity without vision can also be achieved using a conjunctive pose representation . With vision , it is difficult to disentangle the localizing properties of views from knowledge about the arena boundary geometry . For example , it has been shown previously that complex navigation performance patterns within confined arenas may be explained using view-based gradient descent strategies [18] , [19] . These strategies did not require any topological or metric representation of the environment i . e . no “cognitive map” . However , the evidence presented here suggests there may be circumstances where a map-like representation of the local environment is highly advantageous . For example , in darkness and where odor and other allocentric information may be sparse or ambiguous , sporadic contact with natural or artificial boundaries may suffice to maintain place stability for extended periods , until a localizing cue or landmark is detected [42] . In robot navigation research , it has long been known that relatively featureless boundary features such as walls in an office environment can , provided a probabilistic map representation is used , enable a robot equipped with a range sensor to navigate and maintain a correct and stable estimate of its location within the environment [48] . Here , we have applied that principle to situations where boundary information is only available when in close proximity to the boundary , and where local boundary contour does not uniquely identify location . Recently , [44] reported that the cerebellum is important for PI in mice . This conclusion was based on the finding that impaired cerebellar function impaired place field stability in featureless circular arenas . However , we showed that place stability under such conditions is likely to require a boundary map interacting with the PI system . Therefore , the observed reduction in place field stability in cerebellar mutants could be equally explained computationally by a number of different effects involving this map-PI interaction , not just impaired PI . In particular , mutants showed stable place fields in darkness in the presence of a single haptic boundary landmark – suggesting that iPI was not completely impaired . Our results highlight the importance of using quantitative models to determine the computational demands of specific tasks such as PI . Further experiments will be required to carefully dissect the contribution of arena boundary information during spatial navigation tasks , given the significant role it can play in maintaining a stable representation of position . An important question which remains to be addressed in future work is how a rodent's navigation system builds a useful representation of the environment . Some biological aspects of this important question has been addressed using blind rats [42] . Without any local landmarks within the circular arena , place fields were not observed . These rats were free to make contact with the boundary wall , but had no memory information , at least initially . Using our existing particle filter model , it is currently impossible to update the iPI estimate of position in the absence of arena memory . Under a Bayesian formulation , this is equivalent to the case where the likelihood term is not available , so the posterior distribution is unchanged . Consequently , the result is similar to pure iPI – where no place field is seen . During exploration , a noisy estimate of pose can in principle be used to build a representation of the boundary . This is the problem of SLAM ( simultaneous localization and mapping ) which has been studied extensively in robotics . One avenue of future investigation may be to combine rodent-inspired SLAM models ( e . g . , [33] ) with a boundary representation to study questions related to the acquisition of a novel boundary map , including optimal movement strategies . Using a simple probabilistic place field model , it was found that emergent firing fields were related to a true ensemble representation of place in complex and unexpected ways . Firstly , the location of a place field in an arena affected the rate of decrease of spatial information in the absence of vision . In these simulations , it was found that fields near the center of circular arenas persisted for longer , preserving a higher amount of spatial information than fields closer to the boundary . Furthermore , the average spike rate varied considerably and may even increase for some time as place destabilization proceeded . Secondly , the arena shape per se , independent of traversable area , influenced place stability without vision , thereby affecting spatial information content of place fields . These results have a number of implications . Firstly , Skaggs information should be interpreted carefully in the context of positional stability . It is undoubtedly a useful quantifier of positional or directional specificity , but may not accurately reflect the true accuracy and precision of the underlying navigation system . We have shown that it is theoretically possible for identical trajectories and spatial representations to give rise to different values of spatial information content , depending on place field location . Secondly , spatial specificity measures like Skaggs information do not distinguish between unimodal and multimodal fields , or spurious results due to extremely low spike counts . The former is confounded by true field splitting [56] , while our simulations suggest the peak spike rate itself may be affected by an interaction between field position and arena shape . Conversely , it is clear that a navigation system with high place stability is computationally able to generate a firing field with stable spatial information content . Less intuitively , high spike rates did not always correlate well with high place stability or high spatial information content . For example , using iPI and arena memory only , the spatial information content for each 8 minute window was approximately 1 bit/spike or higher , while the spike count increased by nearly 40% over the 48 minute period ( Table 1 ) . The latter result could have been interpreted as indicating a stable representation of place – but this was not the case . Similarly , the spatial information content between 32 to 40 minutes without vision was greater for the simulated place field at 10 cm from the arena center , than between 8 to 16 minutes for the simulated place field at 20 cm from the arena center ( Table S3 ) . If these fields were recorded from separate experiments , the result may have been interpreted as indicating differences in the availability and/or use of spatial cues between the experiments , but this was not the case in the simulations . Thirdly , even though we purposefully modelled the situation where visual information was absent , it is likely that animals use non-visual information when visual information is present , leading to redundancy . It is therefore plausible that neural computational demands may vary substantially from one experimental design to another , depending on the degree of information redundancy . Even controlling for total area , we showed that the shape of an arena affected the navigation performance of the same navigation model ( e . g . , Fig . 3D , Fig . S6 , Table 1 vs Table 2 ) . Furthermore , we have shown there are large differences between iPI in open and enclosed spaces ( e . g . , Fig . 3A ) . Consequently , if there is any spatial navigation element to a task in question , either explicitly or implicitly , the available spatial information must be considered carefully , and quantified where possible . Finally , it is important to note that our models did not make explicit predictions about theta phase [e . g . , 59] . There is a systematic relationship between a rat's position within a place or grid field and the theta phase of cell spikes . This raises the possibility that the navigation system has more precise positional information than suggested by the spatial specificity of the entire field . Therefore , the fact that our model predicts spatial information content of over 4 bits/spike in some cases may be due in part to the fact that we have not embedded the positional information in a phase code , which may reduce the apparent spatial information measured in the conventional way . The latter implies that spatial information may need to be considered as a joint function of spike rate and theta phase . Current models of mammalian PI [34]–[37] , [59] usually assume stable allocentric direction information as input . In contrast , our results show that absolute HD information is not sufficient in the presence of drift , whereas angular displacement information is sufficient for effective navigation in darkness . It should be noted that these results do not prove that angular velocity information per se is used , since angular displacement could be inferred from the change in successive absolute HD estimates . However , past experiments have shown that temporary inactivation of the vestibular system disrupts both the location-specific firing of place cells and the direction-specific firing of HD cells despite visual and odor cues being available [60] , [61] . Together with the existence of large populations of AHV cells ( more populous that HD cells in some regions [38] ) , these evidence strongly suggest that angular velocity or acceleration information is critical for optimal function in the rodent spatial navigation system . One avenue of future research is to incorporate angular head velocity models calibrated using vision [57] , [58] into existing models of mammalian PI . Specific effects of noise on the performance of the oscillatory interference and attractor classes of mammalian PI models have been considered previously [62] , [63] . However , it is clear that allothetic information such as boundary geometry must be used in conjunction with PI for accurate localization , rather than relying on PI in isolation . It will be necessary to investigate the cellular and computational basis for acquiring and using a boundary map . One theoretical approach is to extend the existing mammalian PI models to incorporate a boundary map , and determine whether it is possible to achieve stable place/grid fields assuming realistic sensory inputs and error magnitudes . A second approach is to consider the movement strategies and possible cues which a rat may use to acquire a boundary representation in the first place . Two candidate cell types under modeling investigation are the boundary vector cell [30] , [31] , [56] , [64] , [65] and border cell [29] . In conjunction with a position code , boundary neurons may be able to encode an arena boundary shape . The current work focused specifically on situations where allothetic cues were purposefully minimized or removed . Hence it was by design that only boundary cues could reasonably have been expected to provide stable allocentric information . But as [23] stated , “when boundaries are not available , other types of landmarks can be effectively recruited by the mapping system” consistent with [17] . Indeed , our results are restricted to one particular set of scenarios where a “cognitive map” may be the only plausible explanation of biological data , namely place and grid field stability without vision or olfaction . Our systems-level model does not preclude other stable cues from being incorporated if or when they are available ( e . g . , [33] , [66] ) . Indeed , there are scenarios where a modular navigation system may perform as well as any “cognitive map model” ( discussed further in Text S14 ) . Nevertheless , in at least some conditions , our results demonstrate clearly that as separate modules ( e . g . , [13] ) , iPI and boundary information are much inferior to a unified , near-optimal combination of both . Maintaining separation of PI and landmark ( in this case a featureless boundary ) modules as per [13] cannot support stable place or grid fields without vision or olfaction beyond 2 to 3 minutes . Therefore , to explain observed place and grid cell firing properties , it is necessary that PI and landmark information are combined , consistent with Tolman's analogy that inputs are “worked over and elaborated” [1] . We also demonstrated that a near-optimal probabilistic combination of iPI and boundary information is computationally sufficient to generate stable place and grid fields without vision or olfaction . If our interpretations of published experimental data are reasonable , it would be difficult for any model without a cognitive-like map to produce stable place or grid fields without vision . The arguments of [13] that insects do not possess a “cognitive map” cannot be extended to rodents , possibly reflecting general differences in the spatial navigation systems of different animal phyla . An important question is whether the necessity of fusion of iPI and boundary information should be considered as sufficient evidence of a “cognitive map” in rodents . Although this issue is partly one of semantics and definition , we note a number of important points . Firstly , the necessity of information fusion is at odds with the stimulus-response class of models which Tolman used to contrast against the “cognitive map” class of models . Secondly , the fusion of information is consistent with Tolman's notion of a “central control room” in describing a “cognitive map” , where information from various sources is combined in a coherent manner to produce useful output . Thirdly , the fundamental output of PI systems is metric spatial information . If boundary information has to interact in a useful manner with PI information , it must also contain metric information . It is difficult to envisage a useful representation of an arena boundary with embedded metric spatial information , which bears no resemblance to a spatial map . Fourthly , neither fusion of iPI and boundary memory information only , nor intermittent use of a modular navigation system can maintain stable places fields , showing that the use of a “cognitive map” per se is not always sufficient under the conditions described . The nature and degree of information fusion in using a “cognitive map” are important . Finally , it has been demonstrated that it is possible for a “cognitive map” model which combines iPI and boundary information in a near-optimal manner to explain stable place and grid fields without vision or olfaction . The rodent HD system becomes unstable in darkness beyond 2–3 minutes , consistent with the theory that allocentric cues are required to maintain long term stability . Place cells and grid cells in rats can show stable firing fields for over 30 minutes in darkness . We have shown that these results are incompatible with PI or a boundary representation if they are used independently . However , a “cognitive map” model which combines both can support stable place and grid fields in various arenas while using a drifting HD system . This model predicts place and grid field splitting , and under some conditions grid field rescaling , without vision . The results support the utility of a conjunctive pose representation for optimal navigation , and show how rats might be able to navigate effectively in environments where visual and olfactory cues are unreliable or absent . Seemingly featureless boundaries are powerful landmarks , when combined with a PI mechanism , for stable medium to long term navigation . The influence of such powerful landmarks on navigation tasks must not be underestimated in experimental design or data interpretation . | Do animals need “cognitive maps“ ? One of the main difficulties in answering this question is finding a definitive scenario where having and not having a “cognitive map“ result in measurably different outcomes . Many key predictions made by models involving some sort of “cognitive map“ can also be replicated by models without a “cognitive map“ . Here we consider published data on rodents navigating in darkness inside homogeneous arenas . The head direction system becomes unstable within three minutes in darkness , yet place and grid cells have been reported to fire in the same locations for thirty minutes or longer . We show firstly that it is theoretically implausible for path integration alone to maintain a stable positional representation beyond three minutes , given a drifting head direction system in darkness . Secondly , we prove that even assuming perfect boundary knowledge is insufficient to maintain a stable positional representation . Finally , we show in simulated and real arenas that a nearoptimal combination of path integration and boundary representation is sufficient to produce stable positional representations in darkness consistent with published data . The necessity for fusing path integration and landmark information for accurate localization in darkness is both consistent with , and motivates the existence of , “cognitive maps . “ | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biology",
"neuroscience"
] | 2012 | Maintaining a Cognitive Map in Darkness: The Need to Fuse Boundary Knowledge with Path Integration |
Charcot-Marie-Tooth disease ( CMT ) represents a family of related sensorimotor neuropathies . We studied a large family from a rural eastern Canadian community , with multiple individuals suffering from a condition clinically most similar to autosomal recessive axonal CMT , or AR-CMT2 . Homozygosity mapping with high-density SNP genotyping of six affected individuals from the family excluded 23 known genes for various subtypes of CMT and instead identified a single homozygous region on chromosome 9 , at 122 , 423 , 730–129 , 841 , 977 Mbp , shared identical by state in all six affected individuals . A homozygous pathogenic variant was identified in the gene encoding leucine rich repeat and sterile alpha motif 1 ( LRSAM1 ) by direct DNA sequencing of genes within the region in affected DNA samples . The single nucleotide change mutates an intronic consensus acceptor splicing site from AG to AA . Direct analysis of RNA from patient blood demonstrated aberrant splicing of the affected exon , causing an obligatory frameshift and premature truncation of the protein . Western blotting of immortalized cells from a homozygous patient showed complete absence of detectable protein , consistent with the splice site defect . LRSAM1 plays a role in membrane vesicle fusion during viral maturation and for proper adhesion of neuronal cells in culture . Other ubiquitin ligases play documented roles in neurodegenerative diseases . LRSAM1 is a strong candidate for the causal gene for the genetic disorder in our kindred .
Charcot-Marie-Tooth disease ( CMT ) comprises a set of genetically heterogeneous disorders of the peripheral nervous system , affecting motor and sensory function . CMT is the most common inherited neuromuscular disorder , with a wide range of clinical presentations , but as described by OMIM ( 118200 ) , the salient features of CMT include a slowly progressive weakness and atrophy of the musculature , predominantly of the distal lower limb . This weakness often affects the patients ability to walk or run , and eventually can progress to reach the upper extremity . Within the broad group of patients defined clinically , there are various categories of CMT defined by neurophysiological subphenotypes , pathological findings on biopsy , modes of familial transmission , and specific mutated genes identified in individual patients . These criteria have been extensively reviewed in recent literature [1]–[17] . A query of OMIM for genes causing Charcot-Marie-Tooth yields 26 separate entries with allelic variants; the database of inherited peripheral neuropathies notes 31 gene entries for CMT plus an additional 7 described as causing distal hereditary motor neuropathy . Nonetheless , mutations in new genes associated with CMT continue to be reported[18] . The functions of genes whose mutation yields a CMT or closely related motor neuropathy phenotype span a wide range of disparate biochemical activities including structural components of myelin ( PMP22 , P0 ) , a mitochondrial transport and fusion protein ( MFN2 ) , transcription factors ( SOX , EGR2 ) , components of protein degradation pathways ( DNM2 , RAB7 , LITAF ) , tRNA synthetases ( GARS , YARS ) , a nuclear structural component ( LMNA ) and others [19] . Thus , novel CMT genes are difficult to predict through selection of biological candidates for sequencing in unexplained patients . The best approach for identifying the genetic cause of unexplained CMT remains linkage mapping in multiplex families , with adequate statistical power dependent on the mode of transmission , the specifics of pedigree and local population structure . We report the mapping of a novel form of autosomal recessive axonal CMT through homozygosity mapping in an extended consanguineous pedigree of a local founder population . The identified gene appears to play a role in vesicle metabolism , consistent with some other CMT genes .
In the course of clinical work , we ascertained a patient with Charcot-Marie-Tooth disease , most closely similar to subtype AR-CMT2 ( recessive , axonal ) , although this clinical presentation has sometimes been included as a type of CMT4[16] . The index patient noted the gradual onset of weakness around age 20 , particularly affecting his distal lower extremities , but also present in the hands . He noted episodic muscle cramping of extremity and trunk muscles . He lost the ability to run in his early 20s . He denied sensory symptoms . He had erectile dysfunction and urgency of urination , but no other autonomic symptoms or evidence of spasticity . At the time of examination he demonstrated bilateral pes cavus , with marked wasting of distal lower extremity muscles and mild wasting of hand intrinsic muscles . Fasciculations were present in upper and lower extremity muscles . In the lower extremities he had grade 4 out of 5 ankle dorsiflexion strength ( MRC scale ) , grade 4 hand intrinsic muscle strength and other muscles were grade 5 . He could not walk on either the toes or heels . There was no gait ataxia . Upper and lower extremity tendon reflexes were absent . He had mild loss of sensation on the fingertips and severe loss of sensation in the feet and legs , most markedly to vibration , but also involving proprioception and pain perception . Laboratory investigation demonstrated an elevated serum creatine kinase ( CK ) from 1082 to 1921 U/L ( 18-199 U/L ) . Nerve conduction studies and needle electromyography demonstrated a diffuse sensorimotor peripheral neuropathy . There was no evidence of a primary muscle disorder . The predominant electrophysiological pattern was consistent with axonal degeneration ( see Table S1 ) . Sensory nerve action potentials were small or absent . All of the upper extremity motor nerve conduction velocities were faster than 38 m/s . The ulnar compound muscle action potential amplitude was small and a repeat study 2 years later demonstrated both median and ulnar compound muscle action potential amplitudes were small with normal motor conduction velocities . These are accepted criteria for an axonal CMT [1] . Upper and lower extremity muscles demonstrated ample denervation and partial reinnervation , with fibrillation and reduced recruitment of large motor unit potentials . Denervation of paraspinal muscles indicated axonal degeneration was present at very proximal nerve levels . Temporal dispersion was seen in some motor nerve conductions , but no conduction block , which may be an indication of an element of secondary demyelination , but the predominant electrophysiologic pattern was axonal . The proband is a member of an extended multiply consanguineous family derived from a rural eastern Canadian population isolate; the extended pedigree includes five additional affected individuals with similar suites of symptoms ( Figure 1A ) . Other affected family members exhibited sensory and motor dysfunction with pes cavus . Autonomic symptoms have not been consistently reported . Weakness and wasting has usually been moderate and predominantly in distal lower extremity muscles . The onset of symptoms has usually been in early adult years . One patient was not aware of any difficulties , but had examination abnormalities in his 40's . Some of the affected individuals are able to ambulate into later years , though others have become wheelchair dependent . Sensory symptoms are sometimes not reported , but sensory examination is consistently markedly abnormal , with loss of vibration sense often up to proximal legs and hips . Proprioception loss has been severe in some affecteds with accompanying sensory ataxia . Laboratory abnormalities that are available in only a few patients include mild increased CSF protein and increased serum CK . One patient had significant essential tremor , but that has not usually been reported . When EMG data is available , the pattern is typically predominantly axonal degeneration with only mildly slowed or normal motor nerve conduction velocities and no upper extremity motor nerve conduction velocities slower than 38 m/s . One other patient had evidence of paraspinal muscle denervation , with a normal MRI of the spine , suggesting axonal degeneration at very proximal nerve levels from the neuropathy . Based on transmission of the trait in the pedigree , the genetics are consistent with an autosomal recessive disorder . Given the isolation of the regional population , it seemed likely that all affected individuals in our cohort might be homozygous for the same causal mutation , sharing a chromosomal haplotype around the causal gene . We sampled DNA from six affected patients and related family members . We performed high density genome-wide SNP genotyping of five affected individuals . Formal linkage analysis using a recessive model was not deemed useful , given the highly consanguineous pedigree structure and also the impossibility of obtaining reliable marker allele frequencies for this small subpopulation . Instead , we used the homozygous haplotype ( HH ) method to test for linkage to any of 23 known relevant CMT loci . The HH method is a rapid non-parametric algorithm that utilizes the subset of completely homozygous markers in samples from affected individuals , and looks for consistent loci by excluding regions where affected individuals are homozygous for different alleles of a given SNP [20] , [21] . The method is robust due to the high density of commercial genotyping panels . In this case , HH confidently excluded all of the known relevant CMT loci , under the assumption that all five affected individuals in our pedigree are homozygous for the same causal allele . HH flagged three chromosomal regions as potentially linked , on chromosomes 9 , 15 and 17 ( Figure 1B ) . Subsequently we genotyped additional pedigree members including one more affected , and looked for regions of extended homozygosity shared identical-by-state ( IBS ) in the six affected individuals but not in unaffecteds . As shown in Table 1 , among the longest series of consecutive homozygous SNPs , a region on chromosome 9 appeared as a clear outlier . This region corresponds to that predicted from HH analysis , and extends from rs1324475 at 122 , 423 , 730 Mbp to rs10987845 at 129 , 841 , 977 Mbp . It is interrupted by several single heterozygous SNPs , mostly in one particular sample; these presumably represent false heterozygote genotype calls . In contrast , the potential regions found by HH on chromosomes 15 and 17 were not homozygous in all six affected individuals when all marker data was considered . The likely linked interval is 7 . 42 Mbp in size on chromosome 9 , and includes 84 RefSeq annotated genes , including a cluster of 14 olfactory receptor genes which were not considered likely candidates . We prioritized genes likely to have neuronal or neuromuscular function based on manual review . In all we sequenced 314 coding exons of 18 genes ( HSPA5 , DENND1A , RABGAP1 , RAB14 , STXBP1 , DNM1 , SPTAN1 , DAB2IP , LHX2 , TOR1A , GSN , LHX6 , LMX1B , CDK9 , CDK5RAP2 , FPGS , SH2D3C , LRSAM1 ) , until we observed a particular homozygous variant in the gene LRSAM1 ( Figure 2A ) . This variant changes a coding exon consensus splice acceptor AG dinucleotide to an AA . There are three RefSeq annotated isoforms of LRSAM1 , differing in the 5′ noncoding region , generating transcripts of either 25 or 26 exons . All three splice forms predict the same open reading frame; the variant identified in our patients lies in the penultimate coding exon , either 24 ( isoforms 1 , 2 ) or 25 ( isoform 3 ) . The variant was found homozygous in all six affected individuals , and either wild type or heterozygous as expected among sequenced parents and siblings ( Figure 1A ) . This variant is not present in dbSNP build 130 which includes 2 million novel SNPS recently submitted by the 1000 Genomes project , nor was it detected in any of 150 local control ( a mix of anglo- and franco-phonic individuals ) or 96 CEPH Caucasian control samples , totalling almost 500 control chromosomes . No other homozygous coding variants were detected by sequencing this set of candidate genes . The variant in question changes the consensus splice acceptor site . We tested three splice site prediction programs ( Berkeley Drosophila Project , NetGene2 and SplicePort ) to see whether they were sensitive to the alternative site used in the homozygous patients . All three programs correctly predicted the bona fide splice acceptor site in the wild type sequence . The Berkeley tool failed to predict the alternative AG two nucleotides internally in the mutant sequence , while NetGene2 and SplicePort predicted this acceptor site though with low confidence . We were able to test directly whether splicing of the exon was altered , using total RNA extracted from a fresh blood sample from one affected patient ( 1702 ) . By qualitative RT-PCR , we saw a product of the appropriate size in both a control sample and the affected patient sample , at roughly equivalent intensities ( d . n . s . ) Although the resolution of the electrophoresis was much less than single nucleotide , sequencing of the sample product from the affected patient showed that splicing was to the next AG directly following the true acceptor site , two bases into penultimate exon 24 ( or 25 as per isoform 3 ) ( Figure 2B ) . This causes an obligatory frameshift , leading to an altered open reading frame and premature truncation of the protein after 643 ( out of 723 ) residues in all three spliced isoforms . The effect of this change on protein expression was tested directly by Western blot using EBV-transformed B-lymphocyte cell lines ( B-LCL ) derived from a healthy control and from one of the affected CMT patients . While a single strong band was detected by the anti-LRSAM1 antibody in the control B-LCL ( molecular weight approximately 78 kDa ) , no protein was detected in the B-LCL derived from the CMT patient ( Figure 2C ) . Either the truncated protein is rapidly degraded , or else is rendered non-reactive with our antibody . In either case , the result is most likely to be a significant loss-of-function of the gene product , although unusual gain-of-function effects of a truncated protein can be imagined ( though these might be expected to behave in a dominant not recessive fashion ) . LRSAM1 , leucine rich repeat and sterile alpha motif containing 1 , is predicted to be an E3 type ubiquitin ligase [22] . It is also known as TAL ( TSG101-associated ligase ) and RIFLE . TSG101 itself is a tumor suppressor gene , with a reported role in maturation of human immunodeficiency virus , and LRSAM1 is implicated in its metabolism directly by polyubiquitination . TSG101 is involved in retroviral vacuolar budding . Interestingly , another TSG101-ubiquitinating ligase is known , ( Mahogunin , or MGRN1 ) , for which knockout mice exist and exhibit a neurodegenerative phenotype . Moreover , the known CMT gene LITAF , also called SIMPLE , interacts with mouse ubiquitin ligase gene product NEDD4 [23] , also potentially with TSG101 [24] , and may itself be an E3 ubiquitin ligase [25] , These related findings support the interpretation that mutation of LRSAM1 is probably causal in our patients . It remains to be determined whether the pathogenic effects of mutations in these protein degradation pathway genes act directly via specific neuron-specific proteins ( such as PMP22 ) or more generally through decreasing cell viability . The currently recommended diagnostic paradigm for Charcot-Marie-Tooth entails a complex flow chart combining clinical , familial and molecular genetic analyses [3] . While this approach makes sense when DNA sequencing technologies are cost-limiting , this mixed paradigm could soon be replaced by a more comprehensive and pre-emptive molecular analysis . With the advent of whole genome reagents such as all-exon hybridization capture oligonucleotide libraries , together with the tremendous cost-reductions in DNA sequencing using next-generation nanotechnologies , it should soon be feasible to sequence either entire patient genomes , or entire exomes , for less than the cost of traditional Sanger-based fluorescent capillary sequencing of sets of candidate genes [26]–[28] . We envisage an analysis paradigm whereby all patients with a potential genetic diagnosis , across any medical subdiscipline , may first be sequenced to identify likely pathogenic variants , which can then be cross-indexed with clinical parameters to flag likely causal genes . This approach has recently been shown to be feasible in a research context , including detection of a pathogenic variant in a family segregating a known form of CMT [29]–[32] .
Approval for the research study was obtained from the Capital Health research ethics board . Patients were identified in the course of routine clinical ascertainment and treatment of movement disorders in the neurology clinic at the Halifax Infirmary . All sampled family members provided informed consent to participate in the study . DNA was obtained from blood samples using routine extraction methods . Whole-genome SNP scanning was performed at the McGill University and Genome Quebec Centre for Innovation , using the Illumina Human610-Quadv1_B panel . Data were scanned using the Bead Array Reader , plate Crane Ex , and Illumina BeadLab software , on Infinium II fast scan setting . Allele calls were generated using Beadstudio version 3 . 1 with genotyping module . Data are generated in three different output formats , AB , Forward strand , and Top strand ( as defined by Illumina ) . We used AB format for all linkage analyses . Homozygosity haplotype ( HH ) analysis was performed according to the method of Miyazawa [21] . The source code of HH program was modified to customize the format of output . The parameter LARGEGAP defined in the header file , which is used to define large gap of two consecutive SNPs like centromere , was changed from the default value 300 , 000 bp to 400 , 000 bp to accommodate some non-centromere spaces for HumanHap610 genotypes . The revised C source code of HH program was compiled with GNU compiler on a Linux-based operating system Fedora . HH analysis requires a SNP annotation file , which includes SNP name , physical coordinates , genetic distances , and minor allele frequencies . The SNP annotation file provided by HH software is for the Affymetrix 500K GeneChips Human Mapping Array Set . The HH format annotation of Illumina HumanHap610 for CEPH population was created from the SNP annotations downloaded from Illumina website . The genetic distances of SNPs with empty value , inconsistent value , or zero were interpolated according to the physical coordinates of their flanking SNPs . HH analysis was performed with a cutoff value 3 . 0 cM . Homozygosity analysis was performed using customized scripts and manual inspection comparing samples from affected and unaffected pedigree members . Annotated coding exons were amplified by PCR using standard methods , and sequenced at Dalhousie University , using Sanger fluorescent sequencing and capillary electrophoresis . Sequence traces were analyzed using MutationSurveyor ( Soft Genetics , Inc . ) Specific primers for amplification of LRSAM1 exons and PCR conditions are provided in Table S2 . EBV-transformed B-LCL cells derived from a healthy subject or CMT patient 1675 were cultured in RPMI with 10% FBS and 1% pen/strep in 5% CO2 . Cells were pelleted and lysed in lysis buffer ( 50 mM Tris-HCL , pH 7 . 4 , 150 mM NaCl , 2 mM EDTA , 0 . 2% Triton X-100 with 1 mM PMSF and protease inhibitor tablet ( Sigma ) added to ice cold buffer immediately prior to use ) . Cells were broken by vortexing for 1 minute . Cell debris was removed by centrifugation at 16000×g for 10 minutes . Protein concentration was determined by the Bradford method ( Sigma ) . Samples were diluted to 6 microg/microL in lysis buffer , then to 2 microg/microL in sample dye ( 125 mM Tris-HCL ph 6 . 8 with 20% glycerol , 4% SDS , 0 . 04% bromophenol blue , 10% 2-mercaptoethanol ) . Samples were heated to 95°C for 5 minutes prior to separation of 50 ug sample on a 7 . 5% SDS-PAGE gel . Benchmark pre-stained protein ladder ( Invitrogen ) was included on the gel . Protein was transferred by wet transfer to methanol-wetted PVDF membrane in transfer buffer ( 25 mM Tris-base , 192 mM glycine ) . Membranes were blocked overnight in blocking buffer ( 5% skim milk powder , 0 . 05% Tween 20 , in PBS pH 7 . 4 ) . Anti-LRSAM1 antibody ( abcam ) diluted 1∶500 in blocking buffer was incubated overnight at 4 degrees . Blots were washed in PBS- ( 0 . 05% Tween 20 in PBS pH 7 . 4 ) 15 minutes plus 3×5 minutes . HRP labelled secondary anti-mouse antibody , diluted 1∶2500 in blocking buffer , was incubated for 1 hour at room temperature . Blots were washed as above . HRP was visualized using SuperSignal West Pico Substrate ( Fisher Scientific ) and exposing to X-ray film for 3-5 minutes . Protein transfer to the gel was confirmed by staining the PVDF membrane with Fast Green . The URLs for the data and analytic approaches presented herein are as follows: Online Mendelian Inheritance in Man ( OMIM ) , http://www . ncbi . nlm . nih . gov/Omim/ UCSC Genome Browser , http://genome . ucsc . edu/ NCBI , http://www . ncbi . nlm . nih . gov/ Database of inherited peripheral neuropathies , http://www . molgen . ua . ac . be/CMTMutations/Home/Default . cfm | Sensory motor neuropathies are diseases of the peripheral nervous system , involving primarily the nerves which control our muscles . These can result from either genetic or non-genetic causes , with genetic causes usually referred to as Charcot-Marie-Tooth ( CMT ) disease after the three clinicians who first described the key diagnostic markers . CMT patients lose muscle function , mainly in their arms and legs , with increasing severity during their lives . There are almost two dozen known genes that can mutate to cause CMT , and these fall into a wide variety of biochemical cellular pathways . We identified a group of patients with CMT from a small rural community , with good reason to suspect a genetic basis for their disease . Using high-throughput mapping and DNA sequencing technologies developed as part of the Human Genome Project , we were able to find the likely mutated gene , which was not any of the previously known CMT genes . Based on its sequence , the gene , called LRSAM1 , probably plays a role in the correct metabolism of other proteins in the cell . Among the known CMT genes , some are also involved in protein metabolism , suggesting that this is a generally important pathway in the neurons that control muscle activity . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"neurological",
"disorders/peripheral",
"neuropathies",
"developmental",
"biology/neurodevelopment",
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2010 | Mutation in the Gene Encoding Ubiquitin Ligase LRSAM1 in Patients with Charcot-Marie-Tooth Disease |
Cryptosporidium infection causes gastrointestinal disease and has a worldwide distribution . The highest burden is in developing countries . We sought to conduct a systematic review and meta-analysis to identify Cryptosporidium risk factors in Low and Middle Income countries ( LMICs ) . Medline Ovid and Scopus databases were searched with no restriction on year or language of publication . All references were screened independently in duplicate and were included if they presented data on at least 3 risk factors . Meta-analyses using random effects models were used to calculate overall estimates for each exposure . The most frequently reported risk factors in the 15 included studies were overcrowding , household diarrhoea , poor quality drinking water , animal contact , open defecation/ lack of toilet and breastfeeding . The combined odds ratio for animal contact was 1 . 98 ( 95%CI: 1 . 11–3 . 54 ) based on 11 studies and for diarrhoea in the household 1 . 98 ( 95%CI: 1 . 13–3 . 49 ) based on 4 studies . Open defecation was associated with a pooled odds ratio of 1 . 82 ( 95%CI: 1 . 19–2 . 8 ) based on 5 studies . Poor drinking water quality was not associated with a significant Cryptosporidium risk , odds ratio 1 . 06 ( 95%CI: 0 . 77–1 . 47 ) . Breastfeeding was protective with pooled odds ratio 0 . 4 ( 95%CI: 0 . 13–1 . 22 ) , which was not statistically significant . Based on the included studies , crowded living conditions , animal contact and open defecation are responsible for the majority of Cryptosporidium cases in LMICs . Future studies investigating Cryptosporidium risk factors should have a good study design and duration , include appropriate number of cases , select suitable controls , investigate multiple relevant risk factors , fully report data and perform multivariate analysis .
Cryptosporidium is a protozoan parasite with a worldwide distribution infecting humans and animals . In high income countries , Cryptosporidium occasionally causes sizable outbreaks due to contaminated water supplies or food sources [1] . In Low and Middle Income Countries ( LMICs ) , cryptosporidiosis is much more prevalent and is associated with a significant burden of gastrointestinal disease . Cryptosporidiosis is highly prevalent in early childhood , with 45% of children infected before the age of 2 [2] . More recent studies showed higher prevalence rates: 77% in slum dwelling Bangladeshi children [3] and 97% in children under 3 years from a birth cohort in Southern India [4] . A study in children under 2 years old estimated 2 . 9 million and 4 . 7 million Cryptosporidium infections in sub-Saharan Africa and South Asia , respectively [5] . Cryptosporidiosis is associated with watery diarrhoea persisting for over 2 weeks . This chronicity increases the vulnerability of children in LMICs and is the result of interplay of immune naivety , malnutrition and HIV infection [2] . Cryptosporidium infection in children in LMICs is associated with malnutrition , stunted growth , and cognitive impairment [6] . Cryptosporidiosis exacerbates malnutrition and is more severe in malnourished subjects [7] . Cryptosporidium is the second cause of severe diarrhoea in children under 5 years old in sub-Saharan Africa and south Asia and the leading cause of mortality in children aged 12–23 months [8] . In immunocompromised people such as HIV-positive and transplant patients , cryptosporidiosis is more severe and could result in high mortality rates [1] . The disease burden in both developed and developing countries is likely to be underestimated , due to a large number of asymptomatic or self-limiting diarrhoeal cases , lack of systematic diagnosis of etiologic agent of diarrhoeal disease and reliance on microscopy for routine clinical detection , which is associated with low specificity and sensitivity . Due to the significant burden of cryptosporidiosis , several studies attempted to elucidate Cryptosporidium transmission pathways and risk factors [1 , 7 , 9–11] . The two Cryptosporidium species causing the majority of human infections are C . parvum and C . hominis . The former is transmitted mainly through a zoonotic cycle between humans and animals , while the latter is predominately anthroponotic . The main Cryptosporidium risk factors were summarised in three reviews [7 , 10 , 11] and are related to drinking contaminated water , contact with infected animals or humans ( particularly children ) , consumption of contaminated food , recreational use of contaminated water and travel to disease endemic areas . However , these reviews focused on developed countries and one on the USA . To our knowledge , no such review of cryptosporidiosis risk factors in LMICs exists . Therefore , we attempted to address this gap and conducted a systematic review and meta-analysis of higher quality studies investigating risk factors for Cryptosporidium infection in LMICs .
The methodology and reporting were in accordance with the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” ( PRISMA ) ( S1 File ) . Medline Ovid and Scopus databases were searched with no restriction on year or language of publication up to 12th January 2017 . The search strategy was limited to title/ abstract/ keyword using the following MeSH terms/ keywords: ( Cryptosporidium OR cryptosporidiosis ) AND ( risk factor OR case control OR cohort OR infection OR sporadic OR prevalence ) . Reference lists from relevant papers were screened for additional eligible articles . All references were screened by title and abstract independently in duplicate by MB and EK . Studies investigating Cryptosporidium transmission and risk factors were considered for full text analysis and data extraction . Eligibility disagreements were resolved by discussion . Abstracts without full text or complete results section , such as conference proceedings , were excluded . Only studies from LMICs as defined by the Official Development Assistance ( ODA ) of the Organisation for Economic Co-operation and Development ( OECD ) were included . In order to restrict the analysis to higher quality studies , additional inclusion criteria were applied at full text analysis stage: at least 20 Cryptosporidium infections were reported and the study assessed at least three relevant risk factors . Examples of irrelevant risk factors: age , gender , rural/urban living , stunting , malnutrition ( arguably potential cause and consequence of cryptosporidiosis ) , household income and type of dwelling , as these could not be directly targeted by preventive public health strategies . For each article , the following information was extracted: location of the study , duration , type of study , Cryptosporidium detection method , age range of participants , number of cases , number of controls ( if applicable ) , selection criteria for cases and controls , risk factors ( exposures ) investigated and odds ratios ( or relative risk or hazard ratio ) as reported by the authors or calculated from data presented in the paper ( when available ) . The Newcastle-Ottawa scale ( NOS ) was used to provide a quality assessment score for all included studies [12] . Case control studies were judged across three domains: selection of cases and controls , comparability of cases and controls and ascertainment of exposure . Cohort studies were judged across three domains: selection of cohorts , comparability of cohorts and assessment of outcome . Cross-sectional studies were assessed as per case control studies . Details of subdomains assessed within each criterion are provided in S2 File . A study is awarded a maximum of one star for each subdomain . For this systematic review , only one star was given for the comparability domain as opposed to two possible stars for the traditional Newcastle Ottawa Scale . Therefore , a maximum of 4 stars for selection , 1 star for comparability and 3 stars for exposure/outcome could be awarded , totalling 8 stars if all factors included in the NOS were unlikely to introduce bias . The studies were considered of high quality if NOS score was 6–8 stars , moderate quality for a score of 3–5 stars and of poor quality if the NOS score was 0–2 . The risk factors identified in each study were pooled in a table and categorised . The number of risk factors and the proportion achieving statistical significance were noted . Both univariate and multivariate risk factors were extracted . When at least four papers reported on a particular risk factor , a meta-analysis was performed to calculate a combined random effect odds ratio for this exposure using Reference Manager software ( RevMan ) [13] . Any available ( or calculated ) risk factor was included in the meta-analysis regardless of significance . As the majority of studies reported on univariate estimates of risk factors , only these were considered when calculating pooled odds ratios ( ORpooled ) . Funnel plots generated using RevMan were used to assess publication bias through visual assessment . Population attributable fraction ( PAF ) was calculated using the formula PAF = Pepooled x [ ( ORpooled-1 ) / ORpooled ) ] [14] . Pepooled = proportion of source population exposed to the risk factor , was calculated if the number of exposed cases and total number of cases was available from at least 50% of the studies used to calculate the pooled odds ratio . Pepooled calculations were performed in OpenMeta [Analyst] software [15] , entering the data as untransformed proportions and performing a binary random effects meta-analysis .
Animal contact was investigated in all 15 included studies . The type of animal species ( when available ) is provided in S1 Table . Information needed for meta-analysis could be extracted from 11/15 studies . While the majority of studies found that contact with animals is associated with increased risk of cryptosporidiosis [16–22] , a few reported a protective effect . The combined odds ratio was 1 . 98 ( 95%CI: 1 . 11–3 . 54 ) p = 0 . 02 ( Fig 2 ) . Heterogeneity was substantial with I2 score of 87% . The impact of non-piped drinking water on cryptosporidiosis was investigated in all 15 studies . Meta-analysis was possible for 10 studies . The combined odds ratio was 1 . 06 ( 95%CI: 0 . 77–1 . 47 ) , which was not significant ( Fig 3 ) . This was due to conflicting results between studies , with 6 studies reporting that non-piped water is a risk factor [17 , 18 , 20 , 21 , 23 , 24] , while 4 studies considered it to be protective [19 , 22 , 25 , 26] . Heterogeneity was moderate ( I2 = 33% ) . This risk factor was investigated in 7 studies . The combined odds ratio from 5 studies was 1 . 82 ( 95%CI: 1 . 19–2 . 8 ) p = 0 . 006 ( Fig 4 ) . Heterogeneity was substantial ( I2 = 81% ) . Despite the relatively moderate combined risk associated with open defecation , all studies consistently showed increased cryptosporidiosis risk . Overcrowding was reported as a risk factor for Cryptosporidium infection in 7 studies . The combined odds ratio based on 5 studies was 1 . 37 ( 95%CI: 1 . 07–1 . 75 ) p = 0 . 01 ( Fig 5 ) . Heterogeneity was substantial ( I2 = 72% ) . Only one study reported that overcrowding is protective [23] . Household diarrhoea was a cryptosporidiosis risk factor in 4 studies , all of which were included in the meta-analysis . The combined odds ratio was 1 . 98 ( 95%CI: 1 . 13–3 . 49 ) , which was statistically significant ( Fig 6 ) . Heterogeneity was moderate ( I2 = 38% ) . One study reported that diarrhoea in the household was protective [18] . Breastfeeding was investigated in 10 studies . Meta-analysis was restricted to 5 studies that provided the required information . The combined odds ratio was 0 . 4 ( 95%CI: 0 . 13–1 . 22 ) suggesting a protective overall effect ( Fig 7 ) . However , this was not statistically significant . Heterogeneity was substantial ( I2 = 83% ) . Only one study reported that breastfeeding was conducive to acquiring cryptosporidiosis in infants [21] . For each exposure , publication bias was assessed using funnel plots . All funnel plots had a symmetrical shape suggesting minimal publication bias ( S1 Fig ) . Calculation of population attributable fraction ( PAF ) was possible for 4/5 risk factors . Breastfeeding was protective and therefore no PAF was calculated . Crowding was responsible for 18% of cases ( 95%CI 4–29% ) based on 3 studies ( Table 2 ) . Open defecation was attributable to 17% of cases ( 95%CI 6–25% ) ( based on 4 studies ) , while animal contact accounted for 25% of cases ( 95%CI 5–36% ) ( 8 studies ) . Poor drinking water quality was responsible for 2% of cases ( 95%CI—10% , 11% ) ( based on 8 studies ) .
This systematic review aimed to identify the most frequently reported risk factors for Cryptosporidium infection from LMICs based on good quality studies . Although the search strategy retrieved > 3000 papers , only 15 studies were of acceptable quality warranting inclusion . The pitfall of this strategy is the exclusion of relevant risk factors and decreasing the power of meta-analysis by including fewer studies . Nevertheless , this review identified six risk factors that are likely to be the main drivers of Cryptosporidium infection in LMICs . Animal contact had the highest combined odds ratio 1 . 98 ( 95%CI: 1 . 11–3 . 54 ) , which was statistically significant . This is in accordance with reviews from developed countries where animal contact/ farm visits/ petting zoo visits were significantly associated with acquiring cryptosporidiosis [7 , 11] . Diarrhoea in the household was associated with a similar Cryptosporidium infection risk , pooled odds ratio 1 . 98 ( 95%CI: 1 . 13–3 . 49 ) , which was also statistically significant . Case contact is understandably a risk factor for transmitting any infectious disease and this is relevant in both developed and developing nations . However , as the majority of Cryptosporidium infections are associated with mild self-limiting symptoms in healthy adults and could be asymptomatic in children , the number of diagnosed cases are substantially underestimated , contributing to further Cryptosporidium transmission unless proper hand hygiene and prevention measures are implemented . Similarly , overcrowded living conditions are associated with an increased risk of Cryptosporidium infection , pooled odds ratio 1 . 37 ( 95%CI: 1 . 07–1 . 75 ) . Another person to person transmission pathway is children attending nursery . This risk factor was not assessed in many studies and therefore could not be included in the meta-analysis . This was also the case in developed countries , where only a few studies reported that nursery attendance and changing nappies are risk factors for Cryptosporidium infection [10] . Poor WASH ( Water , Sanitation and Hygiene ) conditions are paramount to the spread of Cryptosporidium and other gastrointestinal infections . The search strategy was not restricted to focus on WASH . Nevertheless , lack of appropriate sanitation/ open defecation was associated with a significant risk of acquiring cryptosporidiosis , pooled odds ratio 1 . 82 ( 95%CI: 1 . 19–2 . 8 ) based on 5 studies . This result is comparable to the systematic review by Speich and colleagues , who reported that lack of sanitation was associated with Cryptosporidium infection risk , pooled odds ratio 1 . 47 ( 95%CI: 0 . 37–5 . 88 ) based on 5 studies [27] . Interestingly , the 5 papers included in our systematic review and the one by Speich and colleagues were different , yet the associated risk was comparable . Our search strategy retrieved the papers included in Speich and colleagues , but these were not included as they considered less than 3 risk factors and/ or the data were missing from the full text . Indeed Speich and colleagues reported that they contacted some of the authors to obtain data that was collected but not analysed/ presented in the full text . This was an issue that we encountered while conducting this systematic review as several authors reported the investigation of several risk factors , which were omitted in the results section . Improving sanitation coverage is one of the aims of the Sustainable Development Goals . Though the number of people practising open defecation globally decreased from 38% to 25% between 1990 and 2015 , there are currently 946 million people lacking sanitation worldwide ( 1 in 8 ) [28] . Open defecation is a clear indicator of extreme poverty and is associated with significant disease burden . In this systematic review , poor drinking water was not associated with Cryptosporidium infection , however , this was not statistically significant . This was due to the contradicting findings of the included studies and wide confidence intervals . We considered the absence of piped water an indicator of poor drinking water quality . However , this is not necessarily true . The microbiological quality of spring and well water could be satisfactory for the majority of time unless contamination events occur . Indeed , many of the papers retrieved by our search strategy highlighted the increased risk of Cryptosporidium infection in the wet season and/or following extreme rain events . Regular consumption of contaminated drinking water , though not recommended from a public health view , could be associated with building protective immunity [29 , 30] . Furthermore , drinking water could be a minor transmission pathway in endemic settings . Indeed , in a quasi-experimental study in India , drinking bottled water was not associated with reduced risk of cryptosporidiosis in children [31] . Breastfeeding ( or lack of ) was investigated in numerous studies that focused on childhood cryptosporidiosis . Breastfeeding was associated with a protective effect , however , this was not statistically significant . The protection potentially conferred by breastfeeding could be due to the passive immunity acquired through ingestion of Cryptosporidium specific antibodies in breast milk [32] . Additionally , bottle feeding was found to increase the risk of cryptosporidiosis [33] , most likely due to one or a combination of the following factors: poor water quality , lack of sterilisation and substandard hand and household hygiene . Indeed , one study found that washing hands before infant feeding was associated with a significant cryptosporidiosis risk , multivariate adjusted odds ratio 5 . 02 ( 95%CI: 1 . 11–22 . 78 ) [34] , which demonstrates the poor quality of water used for drinking and hand washing . The main limitation of this systematic review is the small number of studies included . As strict inclusion criteria were applied , a large number of papers that could have added to the body of evidence were excluded . This was because they had a small number of Cryptosporidium cases , explored less than 3 risk factors ( excluding age , gender , rural/ urban residence , malnutrition ) or reported their results incompletely or inappropriately for inclusion in meta-analysis . This resulted in a small number of studies for each risk factor , which in turn reduced the power of meta-analyses performed . Additionally , this could have inevitably resulted in the exclusion of some other relevant risk factors for Cryptosporidium infection , that are investigated less frequently and/or not in conjunction with well-known transmission pathways . The main limitations of some of the included studies are the small number of cryptosporidiosis cases and poor quality in terms of study design and duration , number of exposures investigated and data reporting . Another shortcoming was that several papers presented risk factors for diarrhoeal diseases in general without categorisation/ sub group analysis for each etiologic agent . Some did not even seek to diagnose diarrhoeal pathogens . This limits the usefulness of such epidemiological studies and hinders the identification of relevant risk factors and the formulation of specific prevention measures . A heterogeneity between the included studies was noted . While , the majority of studies used diarrhoea free , Cryptosporidium negative control groups , some used diarrhoeal subjects that were Cryptosporidium negative . Both sets were undistinguishably included in the meta-analysis , however , combining them would introduce bias in the overall risk associated with each exposure . In summary , this systematic review identified animal contact , diarrhoea in the household and open defecation as the most relevant risk factors associated with Cryptosporidium infection in LMICs . Improving sanitation coverage is one of the Sustainable Development Goals and progress is likely to happen despite the high number of people still practising open defecation globally . Animal contact and case contact/ household diarrhoea are relevant for both developed and developing countries and prevention measures should include awareness campaigns and better hand hygiene . Other relevant risk factors could have been omitted from the systematic review because of the paucity of data and poor quality of several studies . Considering the significant morbidity and mortality of cryptosporidiosis in sub-Saharan Africa and South Asia , especially for under 5 years ( and HIV+ ) , strategies to reduce the prevalence and burden of cryptosporidiosis and other gastrointestinal opportunistic diseases should be prioritised and offered adequate funding . | Cryptosporidium is a parasite that causes diarrhoea and is transmitted through faecal contamination of water and food . Though it occurs in developed nations , it is much more prevalent in developing countries and is associated with high mortality in children under 2 years old . In this review , we looked at published studies on factors that increase the risk of contracting cryptosporidiosis in Low and Middle Income Countries . These factors could be targeted to limit the transmission of the disease . Based on the selected studies , the most important risk factors identified were contacts with animals and presence of infected people in the household . Open defecation was also contributing to the risk of infection by this parasite transmitted through the faecal oral route . Breastfeeding was protective from the infection . Poor drinking water was not responsible for causing the disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"neonatology",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"water",
"resources",
"pathology",
"and",
"laboratory",
"medicine",
"maternal",
"health",
"cryptosporidium",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pediatrics",
"diarrhea",
"protozoans",
"mathematics",
"statistics",
"(mathematics)",
"signs",
"and",
"symptoms",
"women's",
"health",
"gastroenterology",
"and",
"hepatology",
"cryptosporidiosis",
"research",
"and",
"analysis",
"methods",
"natural",
"resources",
"mathematical",
"and",
"statistical",
"techniques",
"marine",
"and",
"aquatic",
"sciences",
"research",
"assessment",
"breast",
"feeding",
"eukaryota",
"diagnostic",
"medicine",
"water",
"quality",
"meta-analysis",
"systematic",
"reviews",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods",
"organisms"
] | 2018 | Risk factors for Cryptosporidium infection in low and middle income countries: A systematic review and meta-analysis |
Schistosomiasis is one of the most important parasitic diseases worldwide , second only to malaria . Schistosomes exhibit an exceptional reproductive biology since the sexual maturation of the female , which includes the differentiation of the reproductive organs , is controlled by pairing . Pathogenicity originates from eggs , which cause severe inflammation in their hosts . Elucidation of processes contributing to female maturation is not only of interest to basic science but also considering novel concepts combating schistosomiasis . To get direct access to the reproductive organs , we established a novel protocol using a combined detergent/protease-treatment removing the tegument and the musculature of adult Schistosoma mansoni . All steps were monitored by scanning electron microscopy ( SEM ) and bright-field microscopy ( BF ) . We focused on the gonads of adult schistosomes and demonstrated that isolated and purified testes and ovaries can be used for morphological and structural studies as well as sources for RNA and protein of sufficient amounts for subsequent analyses such as RT-PCR and immunoblotting . To this end , first exemplary evidence was obtained for tissue-specific transcription within the gonads ( axonemal dynein intermediate chain gene SmAxDynIC; aquaporin gene SmAQP ) as well as for post-transcriptional regulation ( SmAQP ) . The presented method provides a new way of getting access to tissue-specific material of S . mansoni . With regard to many still unanswered questions of schistosome biology , such as elucidating the molecular processes involved in schistosome reproduction , this protocol provides opportunities for , e . g . , sub-transcriptomics and sub-proteomics at the organ level . This will promote the characterisation of gene-expression profiles , or more specifically to complete knowledge of signalling pathways contributing to differentiation processes , so discovering involved molecules that may represent potential targets for novel intervention strategies . Furthermore , gonads and other tissues are a basis for cell isolation , opening new perspectives for establishing cell lines , one of the tools desperately needed in the post-genomic era .
Schistosomes are blood-dwelling digenean trematodes with a complex life-cycle comprising a freshwater intermediate snail host and a mammalian final host . Depending on the schistosome species , adults reside in the intestinal or urinary veins , predominantly of mammals [1]–[4] . Schistosome females produce hundreds of eggs per day , of which a significant proportion fails to pass into faeces ( among others Schistosoma mansoni , S . japonicum ) or urine ( S . haematobium ) to continue the life cycle but instead are dispersed by the blood stream into different organs where they can provoke severe inflammation , granuloma formation , hepatosplenomegaly , and even cancer [5]–[13] . Known as schistosomiasis ( bilharzia ) this infectious disease is considered by the World Health Organisation ( WHO ) as one of the most socioeconomically devastating parasitic disease worldwide , second only to malaria [14] , [15] . Schistosomes are the only trematodes to have evolved separate sexes . Furthermore , a unique phenomenon of their biology is that females that have never been in contact with a male , are sexually immature and drastically smaller in size compared to paired , sexually mature females . A constant pairing contact to a male partner is the prerequisite for the differentiation of the female reproductive organs , which account for most of the significantly increased body size of a paired female [16]–[18] . As the eggs represent the causative pathogenic agents of schistosomiasis , the understanding of processes involved in reproductive organ differentiation , fertilisation , and egg-formation are of fundamental importance for understanding the reproductive biology of this exceptional parasite . Praziquantel ( PZQ ) is the commonly applied drug combating all schistosome species in humans and animals , but is exclusively effective against adult worms [19]–[21] . Due to its use over decades there is an increasing fear of emerging PZQ resistance . Although cases of resistance have not been documented in the clinic yet , first evidence of reduced PZQ susceptibility in patients has been reported , and PZQ resistance can be generated in vitro [22]–[25] . This and the fact that no applicable vaccine is actually in sight emphasises the necessity for further research to find novel strategies combating schistosomiasis . Considering the relevance of this parasitaemia , international genome sequencing projects have been completed with support of the WHO and other organisations [26]–[29] . As schistosome research has moved into the post-genomic era , numerous ( sub- ) transcriptome studies [30]–[45] as well as ( sub- ) proteome studies [46]–[53] have been initiated to functionally analyse tissue- and stage-specific gene transcription and expression to identify novel candidates for drug and vaccine development [33] , [54]–[58] . These initiatives were paralleled by proteomics and glycomics to gain deeper insights in gene expression and regulation [59]–[62] . With respect to post-genomic studies , finally , the establishment of schistosomal cell lines has been recognized as a desirable tool . Although progress has been made during the last decades , the generation of permanently dividing cells has not been achieved yet [63]–[68] . In this study we present a novel , straightforward protocol for the isolation of pure and intact testes and ovaries from adult schistosomes . In contrast to other strategies requiring specific equipment and technical know-how to get access to different tissues [34] , [41] , [43] the introduced method is easy to handle , time saving and efficient by providing complete intact organs as well as tissue-specific RNA and protein of high quality and quantity for further analyses . As a proof-of-principle , the first molecular studies analysing the expression of candidate genes demonstrated the value of this approach for detailed characterisation of gene expression , which in one specific case ( SmAQP ) provided first evidence for tissue-specific , post-transcriptional regulation .
All animal experiments have been performed in accordance with the European Convention for the Protection of Vertebrate Animals used for experimental and other scientific purposes ( ETS No 123; revised Appendix A ) and have been approved by the Regional Council ( Regierungspraesidium ) Giessen ( V54-19 c 20/15 c GI 18/10 ) . A Liberian strain of Schistosoma mansoni was maintained in Biomphalaria glabrata as intermediate host and in Syrian hamsters ( Mesocricetus auratus ) as final host [69] . Adult worms were obtained by hepatoportal perfusion at day 42 post infection . Unisexual worm populations were generated by monomiracidial intermediate-host infection as described elsewhere [70] . Adult worms were transferred to Petri dishes of 60 mm diameter size containing 4 ml M199 medium ( supplemented with 10% Newborn Calf Serum ( NCS ) , 1% HEPES [1 M] and 1% ABAM-solution [10 , 000 units penicillin , 10 mg streptomycin and 25 µg amphotericin B per ml] ) in groups of 20 couples , 25 males , or 50 females per Petri dish and kept in vitro at 37°C and 5% CO2 . Immediately before processing , couples were separated by repeated pipetting and the use of featherweight tweezers . Adult males or females ( 50–60 individuals each ) were transferred separately into round-bottomed 2 ml-reaction vessels and washed twice with 2 ml of non-supplemented M199-medium at room temperature ( RT ) . After removal of medium and addition of 500 µl of tegument solubilisation ( TS ) -solution ( 0 . 5 g Brij35 ( Roth ) , 0 . 5 g Nonidet P40-Substrate ( Fluka ) , 0 . 5 g Tween80 ( Sigma ) , and 0 . 5 g TritonX-405 ( Sigma ) per 100 ml PBS ( 137 mM NaCl , 2 . 6 mM KCl , 10 mM Na2HPO4 , 1 . 5 mM KH2PO4 in DEPC-H2O , pH 7 . 2–7 . 4 ) ) the reaction vessels were incubated at 37°C and 1 , 200 rpm in a thermal shaker ( TS-100 , Biosan ) for 5 min to solubilise the tegument in order to make the sub-tegumental musculature accessible for further processing . This step was repeated once ( females ) or twice ( males ) followed by three times washing with 2 ml of non-supplemented M199-medium at RT to remove most of the detergents . Following removal of medium , the musculature consisting of outer circular muscles and inner longitudinal muscles was digested by protease treatment . To this end , elastase Type IV from porcine pancreas ( Sigma , #E0258 ) was freshly dissolved in non-supplemented M199-medium to a final concentration of 5 units/ml and 500 µl added to each sample . Male- and female-containing reaction vessels were incubated with slight agitation ( 600 rpm ) in a thermal shaker at 37°C for approximately 30–40 min , and the worms swirled up manually every 5 min . Progress of protein digestion was monitored by microscopic inspections of 20 µl aliquots . The appropriate time point to stop the reaction was achieved when the medium turned opaque and the female worms were fragmented , but not completely digested . At the same time the male worms appeared as a conglomerate of several flabby individuals . Additionally , some liberated reproductive organs were observed within these aliquots . Testes and ovaries were identified by their characteristic grape-like and peach-like shapes , respectively . After addition of 1 ml non-supplemented M199-medium to each sample the whole content was decanted into Petri dishes of 60 mm diameter size . To completely harvest worm fragments/organs the vessels were rinsed three times with 1 ml of non-supplemented M199-medium . For quality inspection and following organ isolation , the digested worm batches were analysed under an inverted microscope . Most of the intact organs were liberated and ready to be harvested by pipetting . Remaining testes within worm-carcasses were set free by repeated pipetting ( 200 µl-tip ) . Ovaries surrounded by residual parts of the body wall were liberated in a similar fashion . For further purification of liberated testes and ovaries , the organs were collected with a 10 µl-pipette and transferred into 30 mm Petri dishes each containing 2 ml of non-supplemented M199-medium . If indicated , this step was repeated until the organs were completely free of any residual worm fragments or other cellular material . Finally , the organs were collected using a 10 µl-pipette and transferred into a 1 . 5 ml-vessel , and for concentration the testes and ovaries were centrifuged for 5 min at 1 , 000 g and 1 min at 8 , 000 g , respectively . The supernatant was carefully removed by pipetting and the organs frozen in liquid nitrogen before storage at −80°C for further RNA and protein isolation . With some practise the whole procedure takes approximately 1 . 5 hours . The use of freshly isolated adult worms is essential for the success of the described method as organ isolation failed with frozen worms . A schematic work flow is provided as supplementary Figure S1 . Untreated adult control worms from in vitro culture and TS solution-treated worms with removed tegument ( Figure 1 ) were washed three times with 2 ml of non-supplemented M199-medium and once with 2 ml PBS to remove Ca2+-ions . Subsequently , the worms were fixed in EM-fixative ( 2 . 5% glutaraldehyde , 4% formaldehyde in phosphate buffer ( 0 . 1 M final concentration , pH 7 . 2 ) over night ( o/n ) , washed in several changes of buffer and then dehydrated through a graded series of increasing acetone concentrations to minimise shrinkage . They were critical-point dried , mounted on stubs and sputter coated with gold/palladium before viewing in a Jeol JSM-6490LV scanning electron microscope operating at 5 kV . Viability of gonad tissue-containing cells was analysed by Trypan Blue staining . To this end , freshly isolated testes and ovaries were carefully resuspended in 50 µl Trypan Blue solution ( 0 . 4% w/v , Sigma ) in an Eppendorf-tube and incubated at RT for 5 min under slight agitation . The organs were sedimented by brief centrifugation at 1 , 000 g and 35 µl of the supernatant carefully removed by pipetting . After resuspension of the organs the residual 15 µl were transferred onto a microscope slide , covered with a cover slip and immediately analysed under the light microscope ( CX21 , Olympus ) . Images were acquired with a digital camera ( SC30 , Olympus ) and analysed by CellSens Dimension software ( Olympus ) . Total RNA from adult schistosomes and gonad tissues was isolated using the PeqGOLD TriFast reagent ( Peqlab ) according to the manufacturers' protocol . In brief , five adult males and females as well as 50 testes and 50 ovaries were separately incubated with 500 µl TriFast-solution . The adult worms were mechanically homogenized with a plastic piston . After mixing with 100 µl chloroform and centrifugation for separating organic and aqueous phases , the upper aqueous supernatant , predominantly containing total RNA , was carefully removed to precipitate the RNA by adding 250 µl 2-propanol . In order to drive and visualise nucleic acid precipitation , 35 µg glycogen ( RNase-free PeqGOLD glycogen , Peqlab ) was added per 250 µl 2-propanol . Following incubation o/n at −20°C the RNA was concentrated by centrifugation ( 16 , 000 g ) , washed ( by adding 500 µl 70% EtOH followed by another centrifugation step ) , and dried at RT . Finally , each RNA pellet was dissolved in 10 µl DEPC-H2O . RNA quality and quantity were checked by electropherogram analysis employing the BioAnalyzer 2100 ( Agilent Technologies ) . In brief , 1 µl of resuspended RNA was loaded on an Agilent RNA 6000 Nano Chip according to the manufacturers' instructions and analysed using the device setting “EukaryoteTotal RNA Nano assay” . Synthesis of cDNA was performed using the QuantiTect Reverse Transcription Kit ( Qiagen ) and 500 ng of total RNA per reaction . The preceding gDNA-wipeout step to eliminate residual gDNA as well as the following cDNA-synthesis using the RT-primer mix ( included within the kit ) , which consisted of random hexamers and oligo dT-primers , were performed according to the manufactures' protocol . 1 µl of a 1∶40-dilution of each cDNA-sample was used in a standard PCR of 25 µl total volume ( 1× reaction buffer B: 80 mM Tris-HCl , 20 mM ( NH4 ) 2SO4 , 0 . 02% w/v Tween20 , 2 . 5 mM MgCl2 , 200 µM dNTPs , 400 nM of each primer ( Table 1 ) and 2 . 5 units Fire-Pol taq polymerase ( Solis BioDyne ) ) . PCRs were performed in a MasterCycler ( Eppendorf ) under the following conditions: 95°C for 2 min , followed by 35 cycles of 95°C for 30 sec , 60°C for 30 sec , 72°C for 30 sec and a final extension step of 72°C for 6 min . Adult schistosomes ( 50 males and 100 females ) , 280 testes , or 150 ovaries were separately washed once with 2 ml of non-supplemented M199-medium and PBS . 500 µl of 2× SDS sample buffer ( 200 mM Tris/HCl pH 6 . 8 , 6% SDS , 10% β-mercaptoethanol , 20% glycerol , 20 mM pyrogallol , 1 tablet protease inhibitor cocktail ( Roche ) ) and 100 µl of 1× SDS sample buffer were added to the adults and the reproductive organs , respectively . Worm samples were additionally sonicated 3–5 times with intermittent cooling until complete disruption . Samples were denatured at 100°C for 10 min and centrifuged for 10 min at 13 , 000 g . The supernatant was transferred to a fresh vessel and stored at −20°C . Protein samples from adults and gonads were diluted 1∶500 and 1∶250 in H20 , respectively . 2 . 5 µl of each dilution was used for protein concentration-determination by the BCA-method ( Pierce ) according to the manufacturers' instructions and re-analysed densitometrically on an SDS-PAGE by comparison to different amounts of a BSA-standard . The quality of extracted proteins was analysed by 13% SDS-PAGE applying 1 . 2 µg proteins per lane followed by silver-staining . In brief , fixation of the gel was performed by slight agitation o/n in fixative ( 50% ethanol , 10% acetic acid , 0 . 0185% formaldehyde ) and afterwards washed twice for 25 min with 50% ethanol . After sensitising with 0 . 02% Na2S2O3×5 H2O for one minute and washing three times with water , the gel was stained ( 0 . 2% AgNO3 , 0 . 02775% formaldehyde ) for 20 min . Subsequently , the gel was washed three times with water and transferred into another clean plastic bowl . Development was achieved by incubation in 6% Na2CO3/0 . 0004% Na2S2O3×5 H2O/0 . 0185% formaldehyde for 3 to 5 min until signals were clearly visible . Following brief washing with water the development was stopped by treatment in 12% acetic acid/44% ethanol for 10 min . Prior to storage in 1% acetic acid , the gel was washed three times with water for 10 min and scanned . Pools of 100 males and 150 females were transferred into round-bottomed 2 ml-reaction vessels by pipetting and washed once with 2 ml of non-supplemented M199-medium and TS-solution at RT in order to remove most of the tegument-attached serum and host proteins . Subsequently , the worms were treated 6× with 500 µl TS-solution at 37°C and 1 , 200 rpm in a thermal shaker for 5 min to completely solubilise the tegument . The tegument protein-containing fractions were pooled ( gender-separated ) and precipitated by the chloroform/methanol method [71] . In brief , 1 . 4 ml of TS-solution supernatants were transferred into a 15 ml Corex-glass tube and mixed successively by vortexing with 5 . 6 ml methanol , 1 . 4 ml chloroform , and 4 . 2 ml H2O . After centrifugation for 10 min at 14 , 000 g , the upper aqueous phase was carefully removed , 4 . 2 ml methanol added to the bottom phase , vortexed , and centrifuged for 2 min at 14 , 000 g . Lastly , methanol was removed as much as possible without affecting the pellet , the precipitated proteins dried under vacuum , and finally resolved in 500 µl 2× SDS-sample buffer . Protein concentration was determined by the BCA-method as described before . 15 µg protein of each sample were separated by 13% SDS-PAGE and blotted onto a nitrocellulose membrane using a tank blot system ( Roth ) . After washing the membrane with PBST ( 1× PBS containing 0 . 1% Tween-20 ) , blocking was done with 1× RotiBlock ( Roth ) at RT for 30 min . The membrane was horizontally cut into four parts corresponding to the with respect to the size of the different target proteins . Subsequently , the strips were probed separately with the appropriate diluted rabbit-derived anti-sera [72]–[75]: SmSPRM1hc ( Permease 1 heavy chain , 72 kDa , 1∶600 ) , SmHSP70 ( Heat shock protein 70 , 70 kDa , 1∶20 , 000 ) , SmAQP ( Aquaporin , 33 kDa , 1∶600 ) , and SmFKBP12 ( FK506-binding protein , 12 kDa , 1∶3 , 000 ) o/n at 4°C . After washing three times with PBST for 15 min , the membranes were incubated with horseradish peroxidase ( HRP ) -conjugated goat anti-rabbit immunoglobulin G ( IgG ) diluted 1∶10 , 000 for one hour at RT . The strips were washed three times with PBST for 15 min and detection was performed by Enhanced Chemiluminescence ( Pierce ECL Western Blotting Substrate , Thermo Scientific ) and exposure to X-ray films ( Kodak BioMax Light film ) .
In order to get access to different tissues , especially to the reproductive organs , a novel protocol was developed consisting of a combined detergent/protease treatment . Due to the morphological structure of trematodes , the first and crucial step is the removal of the surface membrane complex ( subsequently referred to as tegument ) appearing as a heptalaminate structure being composed of an outer trilaminate membrane forming the membranocalyx and a subjacent normal plasma membrane ( [76]–[78] , Figure 1A ) . The syncytial tegument of the adult schistosome worms is physiologically highly active and constitutes a strong and extremely resistant barrier against external influences . A combination of four different non-ionic and non-denaturing detergents ( Brij35 , Nonidet P40 , Tween80 , and TritonX-405 ) allowed the complete solubilisation of male and female teguments without destroying the integrity of the worm ( Figure 1B ) . As confirmed by SEM-analyses , after tegument removal the musculature consisting of the outward circular muscles and the inward longitudinal muscles represented the outer surface of the remaining worm carcasses . The proteinaceous musculature was carefully digested by elastase , which specifically hydrolyses elastin , a protein component of elastic fibres [79] . Digestion led to the degradation of the worm carcasses and thus to the release of intact reproductive organs and cells of different origin ( Figure 2A ) . Testes and ovaries were easily identified by their characteristic grape-like and peach-like shapes , respectively , and further purified by sequential transfer ( once up to several times ) into new medium by careful pipetting ( Figure 2B ) . The content of the testicular lobes appeared granular and homogenous , whereas the ovaries of mature females appeared in-homogenous as expected , containing immature oogonia in the small anterior and mature primary oocytes in the bigger posterior part , respectively [80] . Depending on the quality of preparation , the majority of testes were liberated consisting of 6–9 testicular lobes with a diameter of 90–100 µm per lobe ( Figure 2A ) . With respect to females , over-digestion led to an increased fragmentation of the ovaries into posterior and anterior parts and , therefore , to a decreased yield of intact organs , whereas a low digestion-efficiency due to lower enzyme concentration/activity or time of digestion , led to higher numbers of ovaries surrounded by residual parts of the body wall . The size of isolated mature ovaries was approximately 400 µm in length and 120 µm in maximal width; as expected immature ovaries isolated from unisexual females were much smaller , about 200 µm length and 50 µm width . Average isolation efficiencies for gonads derived from bisexual infections were about 70% , whereas the efficiencies for testes and ovaries from unisexual infections were 70% and >90% , respectively . Finally , depending on further processing , the organs were either concentrated in reaction vessels by centrifugation , frozen in liquid nitrogen and stored at −80°C , or directly used for staining . Furthermore , it was even possible to isolate ootypes from unisexual females and vitellarium tissue from mature females ( Figure 2A , B ) . Freshly isolated testes and ovaries were stained with Trypan Blue to determine the viability of the cells within the isolated reproductive organs . This diazo dye penetrates cell membranes of dead cells exclusively but is not absorbed by living cells due to the selective exclusion by their intact cell membranes . Therefore , dead cells appear as blue-colour structures , whereas living cells appear more translucent , not being stained . The percentage of living cells within ovaries was estimated to be more than 60% , whereas with respect to testes the number of vital cells was found to be slightly lower and estimated by 40–50% ( Figure 3 ) . To determine if the quality of the isolated organs will be sufficient to serve as a source not only for viable cells but also for RNA and proteins , subsequent experiments were performed . Total RNA isolated from adult males , testes , and ovaries was analysed on an Agilent RNA 6000 Nano Chip ( Agilent Technologies ) for its integrity as this is an important prerequisite for further applications such as cDNA-synthesis and RT-PCRs . The analyses by this microfluidics-based system demonstrated that the quality of RNA isolated from reproductive organs was comparable to that obtained from adult male worms as control; no significant degradation of RNA was detected as proven by the integrity of the 18S rRNA shown by the appropriate peaks ( Figure 4 ) . Quantification of total RNA amounts of gonad tissue from bisexual as well as unisexual adults was done with the same system ( Table 2 ) . The average amount of total RNA per ovary derived from bisexual and unisexual individuals was determined to be approximately 26 ng and 0 . 8 ng , respectively . Independent of the pairing-status comparable RNA quantities of 8 ng and 7 ng per testis were determined . These differences were also reflected by the data obtained for whole adult worms of both gender and pairing-status as RNA amounts of males were comparable , whereas unisexual females yielded about 5-times less RNA compared to bisexual females . RNA content of bisexual males compared to females was similar . Total RNA obtained from testes and ovaries as well as from adult couples was used for cDNA-synthesis and subsequent RT-PCRs . To demonstrate tissue-specific transcription , representative target genes ( SmFKBP12 , SmCNA , SmTGFβRI ) were selected that had been reported in former studies to be transcribed and/or translated within testes and ovaries [81] . Further target genes ( SmAQP , SmSPRM1hc , SmNPP-5 ) were shown to be preferentially , but not exclusively localised in the tegument of adult worms [82] , [72] , [83] . Additionally , SmHSP70 was chosen as it had been detected throughout diverse life stages and tissues as well as SmAxDynIC , which was expected to be expressed in testes due to its predicted function in sperm axonemes ( [84] , Table 1 ) . As expected , most of the analysed target genes were transcribed in testes , ovaries , and adult couples , whereas the nucleotide pyrophosphatase/phosphodiesterase SmNPP-5 was not detected in the reproductive organs . Transcription of the amino acid transporter SmSPRM1hc gene was detected in the gonads of both genders providing indication for a function not restricted to the tegument . Interestingly , SmAQP was transcribed in testes but not in ovaries suggesting a role of this transporter for spermatogenesis but not for oogenesis . SmAxDynIC transcripts were found in testes but not in ovaries as anticipated ( Figure 5 ) . The quality of total protein was checked by SDS-PAGE analysis . To this end equal protein amounts derived from adult male and female , testes , ovaries , and precipitated male tegument fraction were separated on a 13% SDS-gel and subsequently silver-stained . All samples analysed showed a protein distribution over a wide molecular weight spectrum , ranging from 10 kDa to more than 250 kDa ( Figure 6 ) . Protein amounts of bisexual and unisexual adult worms as well as of the corresponding reproductive organs were determined by the BCA-method ( Table 2 ) . The average amount of total protein per ovary derived from bisexual and unisexual individuals was determined to be approximately 0 . 4 µg and 0 . 03 µg , respectively . Independent of the pairing-status comparable protein quantities of 0 . 33 µg ( bisexual ) and 0 . 3 µg ( unisexual ) per testis were determined . These differences were also reflected by the data obtained for adult worms of both gender and pairing-status as protein amounts of males were comparable , whereas unisexual females yielded 3 . 3-times less protein compared to bisexual females . Protein content of bisexual males compared to females was approximately 2 . 6-times higher . The proportion of male tegument protein based on the protein amount of one individual worm was about 27 µg . However , although a pre-washing step was performed it cannot be completely excluded that residual amounts of serum and/or host proteins were still present within this sample , and/or that additional proteins from inside the worms were co-extracted . Tissue-specific protein expression of some of the genes previously analysed by RT-PCRs ( Figure 5 ) was investigated also by immunoblotting employing antisera directed against SmSPRM1hc , SmHSP70 , SmAQP , and SmFKBP12 ( kindly provided by Patrick Skelly and Mo Klinkert ) . To this end equal amounts of protein derived from adult male and female , testes , ovaries as well as precipitated male and female tegument fractions were separated on a 13% SDS-gel and transferred on a membrane by electroblotting . Expression of SmSPRM1hc , SmHSP70 , and SmFKBP12 was detected for all samples analysed , although only SmHSP70 showed a comparable strength of expression within every lane ( Figure 7 ) . Thus it served as a quantitative standard . Compared to this , SmFKB12 was more strongly expressed in females than in males . This was also observed for the appropriate tegument fractions , whereas the SmFKB12 expression levels within testes and ovaries seemed to be nearly identical . For SmSPRM1hc , the strongest signals were detected in adults of both genders . Weaker signals were detected for the reproductive tissue samples , and the weakest signals in the tegumental fractions . Whereas SmSPRM1hc seemed to be expressed slightly higher in testes and the male tegument compared to ovaries and the female tegument , it seemed to be slightly more highly expressed in adult females compared to adult males . SmAQP was found in adults and tegument samples of both genders showing minor dominance in male-derived samples , but it was absent from the gonad tissue .
In the light of published schistosome genome sequencing projects , methods to characterise genes of interest have gained importance in the dawn of the post-genomic era . Among other needs , the access to organs and cells for subsequent analyses is one of the desired aims [63] . First approaches towards cell isolation from schistosomes in the past were successfully performed by mincing adult worms under aseptic conditions and in the presence of trypsin/EDTA . This permitted the access to different kind of cell types suitable for cell culture purposes , but intact tissues and inner organs were disrupted due to mechanical forces [67] , [68] , [85] . Former attempts to isolate internal organs by a simple protease digestion were not successful ( Grevelding , personal communication ) . Therefore , we established a novel method comprising a combined detergent/protease treatment . We assumed that the robust outer tegument had to be removed prior to digestion of the proteinaceous musculature by proteases . Indeed , solubilisation of the tegument turned out to be the crucial step as well as determining the appropriate detergents and their concentrations , as e . g . the use of SDS would lead to complete digestion . The combination of four different non-denaturing and non-ionic detergents was found to be optimal . A final concentration of 0 . 5% of each substance proved to be effective and the most gentle , completely solubilising the tegument but ensuring the integrity of the remaining worm body . Basically , the tegument could also be removed by a combined freezing/vortexing procedure [86] , [87] , but this will be lethal for schistosome cells . As for the detergents , the appropriate protease and concentration for muscle digestion had to be determined empirically . Initially , employing trypsin or chymotrypsin in the presence and absence of hyaluronidase or even proteinase K alone failed to digest the worms efficiently . Moreover , a collagenase/dispase-mixture dissolved in Schistosoma culture-medium [88] without addition of NCS appeared to be suitable for female , but not for male adult worms . Finally , elastase was found to be the enzyme of choice optimal for processing adult worms of both genders . However , this enzyme had to be dissolved in non-supplemented M199-medium . Applying the established protocol we succeeded to isolate intact testes and ovaries from adult males and females , respectively . After digestion , females appeared nearly completely fragmented resulting in liberated and intact ovaries as well as fragments , lobes , and huge numbers of released vitelline cells from the vitellaria , which represented mainly the S4-stage according to their yellowish appearance indicating a fully differentiated vitellocyte with lipids and yolk [89]–[91] . Using the same conditions , however , male worms were not digested completely , and appeared as agglomerated flabby carcasses . Nonetheless , testes were released easily by the procedure probably due to a dorsal vulnerability of the anterior male body , where the testes are located . The released reproductive organs were purified and enriched from the crude preparation by pipetting . This resulted in material free from other tissues or cells , which was essential for further subsequent experiments aiming to characterise tissue-specific gene expression . Average isolation efficiencies for testes and ovaries derived from bisexual as well as unisexual infections were shown to be sufficient to obtain enough material for subsequent experiments . Incidentally , also other organs such as intact ootypes from unisexual females as well as vitellarium tissue and gut fragments ( data not shown ) from bisexual females were obtained . Nonetheless , the presented protocol may serve as a basis for improvements to optimize the isolation of tissues other than gonads . We anticipate that the method is also applicable for other schistosome species as well as for further plathyhelminths having a tegument . To assess the quality of the isolated reproductive organs and to judge if they could serve as a source for viable cells as well as intact RNA and proteins , testes and ovaries were stained with Trypan Blue immediately after isolation . Both types of organs contained numerous viable cells , which is a prerequisite for future attempts on establishing schistosomal cell lines . Future isolations of gonad tissues in the presence of anti-apoptotic substances/apoptosis-inhibitors will show if the amount of viable cells can be increased . As a high percentage of cells within the isolated organs were proven to be viable , the question was addressed whether intact RNA could be isolated from this material . To this end total RNA extracted from reproductive organs of both genders and pairing-status as well as from corresponding adult worms were isolated and analysed by a microfluidics-based system . The results demonstrated that RNA isolated from reproductive organs was of comparable quality with respect to RNA obtained from adult males as shown by the integrity of the 18S rRNAs . The 28S rRNAs could hardly be detected , which is due to a small gap region within the 28S RNA molecule leading to the dissociation into two equal sized fragments [92] , [93] . Nevertheless , isolation of reproductive organs from batches of 50–60 adult males or females will result in sufficient amounts of total RNA for further analyses like RT-PCRs and transcriptomics . Quantity and quality of all protein fractions were verified by SDS-PAGE analysis confirming the absence of obvious protease-mediated protein degradation , as the proteins were spanning a broad molecular weight range . Accordingly , proteins of testes and ovaries can be isolated in sufficient amount and quality for further analyses such as immunoblotting and proteomics . Comparison of bisexual males and females revealed 2 . 6-times more protein in males , whereas RNA amounts were similar between both genders . With respect to unisexual adult worms differences for RNA and protein amounts were determined as 5 . 4- and 8 . 5-times , respectively , which represents a 1 . 6-fold difference . These results indicate more post-transcriptional processes in bisexual compared to unisexual females . RNA amounts as well as protein quantities determined for testes derived from bisexual and unisexual males were comparable , which was anticipated with respect to similar sizes and morphology [80] . As expected , the differences regarding ovaries obtained from unisexual and bisexual females were significant . Ovaries derived from bisexual individuals were determined to contain 32-times and 14-times more RNA and protein , respectively , which is also explained by the influence of pairing on growth and differentiation of this organ in females . Deviations between RNA and protein amounts as reflected by the RNA/protein ratios are indicating a divergent transcriptional and translational activity with respect to these organs . Among others this can be explained by maternal transcription and storage of mRNAs [94] needed for subsequent embryogenesis and miracidial development within the egg . The RNA/protein ratio is generally considered to reflect the growth rate of cells indicating the level of synthesis activities . For example , this ratio was approximately 2 . 6-times higher in bisexual females than in bisexual or unisexual males indicating an increased RNA synthesis and cellular activity in paired , mature females . It was also observed that unisexual females , which are smaller having a much lower amount of total RNA and protein , have also a lower RNA/protein ratio , and thus probably less cellular activities . Whereas testes of bisexual and unisexual males have similar RNA and protein content , immature ovaries of unisexual females are smaller and contain lower amounts of total RNA and protein compared to mature ovaries present in bisexual females . Moreover , it is interesting to note that the RNA/protein ratio in ovaries of bisexual females is 3-times higher as in the corresponding complete worm , indicating a high synthesis activity within the ovary . In small ovaries isolated from unisexual females , this ratio is only 2-times more than in the complete organism , showing lower activity in the ovary of immature females . To demonstrate the applicability of gonad-derived material for subsequent analyses , RT-PCRs and immunoblots were performed . To this end several target genes/proteins were examined that had been characterised in previous studies . SmHSP70 was detected with equal strength both on the transcription and expression level in the gonads of both genders and in all further samples analysed revealing the omnipresence of this molecule as shown in former studies [34] , [35] , [41] , [47]–[49] , [95] , [96] as well as suggesting that HSP70 is ubiquitously expressed serving as a sample loading control for immunoblot-analyses . SmFKBP12 and SmTGFβR1 as members of the TGFβ-signalling pathway have been shown before to be transcribed among others in the ovary but neither in the testis nor the sub-tegumental cells bodies of adult schistosomes by in situ hybridisation experiments [81] . With respect to the aforementioned study , SmFKBP12 expression was detected in the tegument of both genders , the ovary but not in testes by immunolocalisation . Contrary to these findings we demonstrated transcription and translation of SmFKBP12 as well as transcription of SmTGFβR1 also in the testes confirming TGFβ-signalling in the gonads of both genders . As FKBP12 has been reported to interact with the protein phosphatase calcineurin ( CN ) , tissue distribution of the CN-subunit A ( CNA ) was also analysed previously by immunolocalisation [81] demonstrating SmCNA expression in the tegument and parenchyma of both genders as well as in the testes but not in the ovary . Again our own results show SmCNA-transcripts in the gonads of males and females . The partial discrepancies to the former results ( summarised in Table 3 ) is explained by the detection limit of the in situ-hybridisation and immunolocalisation method as RT-PCR and immunoblot-analyses performed on the sub-transcriptomic and sub-proteomic level dramatically increases sensitivity regarding low abundantly occurring transcripts and proteins . Axonemal dynein intermediate chains have been described to function in cilia and flagella as well as in sperm axoneme assembly being important for spermatogenesis and fertility [84] , [97] . Accordingly , SmAxDynIC transcripts were detected in testes but not in ovaries by organ-specific RT-PCRs and consequently could serve as a testes-specific marker . Finally , three genes whose translation products had been shown before to be localised in the tegument of adult schistosomes were analysed as summarised in Table 3 . SmNPP-5 transcription was neither detected in testes nor in ovaries by organ-specific RT-PCRs . This is consistent with previously published results showing the presences of SmNPP-5 predominantly in the tegument and at lower levels in internal tissues suggesting that SmNPP-5 is closely associated with the new tegument surface generation after cercarial penetration [83] . SmSPRM1hc is widely distributed throughout adult male and female worms as determined by immunolocalisation and is involved in the import of diverse amino acids [72] . Unfortunately , sections from adult couples analysed in the aforementioned study did not encompass reproductive organs . RT-PCRs performed on testes- and ovary-specific cDNA , however , revealed the transcription of SmSPRM1hc within both reproductive organs expanding knowledge about the distribution of this transporter and , furthermore , providing evidence for the importance of amino acid uptake also in the gonads . Consistently , SmSPRM1hc was detected by immunoblot-analyses to be expressed in the reproductive organs but also in the tegument of both genders . However , strongest expression was found in adult males and females indicating that SmSPRM1hc is widely expressed in many other tissues . Aquaporins ( AQPs ) are small integral membrane proteins primarily involved in osmoregulation by transporting water across cell membranes . A subgroup of this protein family is additionally capable of transporting glycerol and therefore called aquaglyceroporins [98] . SmAQP , a type 3/9 aquaglyceroporin , was detected strongly in the tegument of both genders with slight predominance within the female fraction , which corresponded well with the results for whole control adult males and females . These results confirmed former findings that SmAQP is most strongly expressed in the tegument of 2-day and 7-day cultured schistosomula [74] as well as in the tegument of adult schistosomes where it was stronger for males than for females [82] . Furthermore , tegumental expression of a type 3/9 aquaglyceroporin was also supported by proteomic approaches [48] , [49] . SmAQP was characterised to be capable of transporting water , mannitol , fructose , and alanine but not glucose , suggesting its important role in nutrient uptake and waste metabolite excretion [74] , [82] . Immunolocalisation data with respect to the reproductive organs were missing in the aforementioned studies , and we showed that SmAQP translation products could neither be detected in testes nor in ovaries . In contrast , however , our RT-PCR results demonstrated the presence of SmAQP transcripts in testes but not in ovaries , which could be explained by post-transcriptional regulation and/or “leaky transcription” . Conclusively , our data emphasise former results [74] , [82] that SmAQP is predominantly expressed and functional in the tegument but not in testes and ovaries . Nonetheless , indications for the existence and function of AQPs of both subgroups within the male reproductive systems in vertebrates but also in invertebrates have been obtained in the past . In rats , AQPs are present in germ cells as well as other tissues within the male reproductive tract and involved in the maturation of germ cells , the early stage of spermatogenesis , and in the cytoplasmic condensation occurring during differentiation of spermatids into spermatozoa [99]–[103] . AQPs were also identified in reproductive tissues of invertebrates indicating similar functions , as the expression of an aquaglyceroporin AQP3 homologue has been located in the seminal vesicle and vas deferens of Caenorhabditis elegans [104] . However , it cannot be excluded that AQPs of type1/2 exclusively transporting water might play a role in reproductive tissues of schistosomes , as AQPs homologous to AQP1 and AQP2 were shown to be expressed in the epithelial lining of ovary and testes in the trematode Fasciola gigantica [105] . Basically , tissue containing RNA and proteins can also be isolated by laser-assisted microdissection ( LMD ) , which was previously shown to be a new method for tissue-specific profiling in schistosomes [34] , [41] , [43] . Downstream applications of LMD like RT-PCR , real time PCR , and microarray analyses are particularly suited because of the possibility of amplifying low amounts of extracted material . However , there is a potential risk of bias in such analyses , since low copy transcripts may not be detected in post-microdissection analyses . Furthermore , this method requires specific laboratory equipment , and the dissection has to be performed very precisely to prevent contamination with unwanted tissue material . Moreover , specimen preparation and preservation of the target material to prevent degradation during tissue processing is also challenging . Proteomic approaches are even more problematic as proteins cannot be amplified , which could be critical with respect to sensitivity . In the current study we present an alternative method for the isolation of schistosome tissues like reproductive organs as a source for RNA and proteins in sufficient amount , quality , and purity for further downstream analyses . The procedure is easy , inexpensive and quickly performed without the need of specific equipment . The amount of material obtained by this method helps to surmount detection limits of other methods like in situ-hybridisation or immunolocalisation as shown for SmFKBP12 . Analyses on the basis of organ-specific cDNA can help to overcome such limitations and are useful for the validation of previously obtained results . Related to this , such sub-transcriptomic and sub-proteomic analyses are recommended to be performed for transcription and expression profiling of genes of interest prior to the performance of target-oriented experiments or prior to the postulation of working hypotheses . The developed protocol allowed also the isolation of very small tissues such as ovaries and ootypes of immature females , which are difficult to obtain by LMD due to the limited amount of accessible material . Future attempts in our group will concentrate on the enrichment of other tissues , e . g . vitelline lobes and the intestine by modifying the current protocol . Furthermore , the described technique opens new perspectives for the isolation of cells , which cannot be achieved by LMD . As cells with stem cell character will be among the isolated material [106] they represent an ideal source for new attempts to establish schistosomal cell lines . These could be of great value for e . g . a constant source of DNA and proteins of schistosomes , a system to express schistosome proteins in a homologous environment , for drug screening experiments and , if transfectable , gene characterisation . | As a neglected disease , schistosomiasis is still an enormous problem in the tropics and subtropics . Since the 1980s , Praziquantel ( PZQ ) has been the drug of choice but can be anticipated to lose efficacy in the future due to emerging resistance . Alternative drugs or efficient vaccines are still lacking , strengthening the need for the discovery of novel strategies and targets for combating schistosomiasis . One avenue is to understand the unique reproductive biology of this trematode in more detail . Sexual maturation of the adult female depends on a constant pairing with the male . This is a crucial prerequisite for the differentiation of the female reproductive organs such as the vitellarium and ovary , and consequently for the production of mature eggs . These are needed for life-cycle maintenance , but they also cause pathogenesis . With respect to adult males , the production of mature sperm is essential for fertilisation and life-cycle progression . In our study we present a convenient and inexpensive method to isolate reproductive tissues from adult schistosomes in high amounts and purity , representing a source for gonad-specific RNA and protein , which will serve for future sub-transcriptome and -proteome studies helping to characterise genes , or to unravel differentiation programs in schistosome gonads . Beyond that , isolated organs may be useful for approaches to establish cell cultures , desperately needed in the post-genomic era . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"functional",
"genomics",
"gene",
"regulation",
"cell",
"differentiation",
"parasitology",
"gene",
"function",
"germ",
"cells",
"developmental",
"biology",
"organism",
"development",
"stem",
"cells",
"molecular",
"development",
"neglected",
"tropical",
"diseases",
"molecular",
"genetics",
"signaling",
"pathways",
"zoology",
"veterinary",
"science",
"infectious",
"diseases",
"signaling",
"in",
"cellular",
"processes",
"veterinary",
"diseases",
"veterinary",
"parasitology",
"fertilization",
"zoonotic",
"diseases",
"organogenesis",
"biology",
"helminthology",
"signal",
"transduction",
"schistosomiasis",
"signaling",
"gene",
"identification",
"and",
"analysis",
"genetics",
"cellular",
"types",
"genomics",
"molecular",
"cell",
"biology"
] | 2013 | Whole-Organ Isolation Approach as a Basis for Tissue-Specific Analyses in Schistosoma mansoni |
Over the last 20-80 million years the mammalian placenta has taken on a variety of morphologies through both divergent and convergent evolution . Recently we have shown that the human placenta genome has a unique epigenetic pattern of large partially methylated domains ( PMDs ) and highly methylated domains ( HMDs ) with gene body DNA methylation positively correlating with level of gene expression . In order to determine the evolutionary conservation of DNA methylation patterns and transcriptional regulatory programs in the placenta , we performed a genome-wide methylome ( MethylC-seq ) analysis of human , rhesus macaque , squirrel monkey , mouse , dog , horse , and cow placentas as well as opossum extraembryonic membrane . We found that , similar to human placenta , mammalian placentas and opossum extraembryonic membrane have globally lower levels of methylation compared to somatic tissues . Higher relative gene body methylation was the conserved feature across all mammalian placentas , despite differences in PMD/HMDs and absolute methylation levels . Specifically , higher methylation over the bodies of genes involved in mitosis , vesicle-mediated transport , protein phosphorylation , and chromatin modification was observed compared with the rest of the genome . As in human placenta , higher methylation is associated with higher gene expression and is predictive of genic location across species . Analysis of DNA methylation in oocytes and preimplantation embryos shows a conserved pattern of gene body methylation similar to the placenta . Intriguingly , mouse and cow oocytes and mouse early embryos have PMD/HMDs but their placentas do not , suggesting that PMD/HMDs are a feature of early preimplantation methylation patterns that become lost during placental development in some species and following implantation of the embryo .
In eutherian mammals the placenta plays a vital role in not only the transfer of nutrients and waste between mother and offspring but also as a protective layer between the maternal and fetal immune systems during fetal development . Despite this , the gross morphologies and cellular characteristics of the maternal/fetal interface are quite diverse [1 , 2] and have undergone multiple instances of both divergent and convergent evolution [3] . Marsupials also have an extraembryonic membrane ( EEM ) that , although short-lived , is also important for nutrient exchange , originates from a trophectoderm layer in the early embryo , and has been argued to be a true placenta [4 , 5 , 6 , 7] . DNA methylation is essential for proper embryo and placenta development . The offspring of Dnmt3a conditional knockout mothers die in utero by E11 . 5 [8] . The offspring of Dnmt3L null mothers die at E9 . 5 from placental abnormalities and/or imprinting defects [9 , 10 , 11] . Another DNMT important for normal placenta development is DNMT1o , an oocyte-specific isoform of DNMT1 that is present in eutherians as well as metatherians [12] . Loss of Dnmt1o in mice results in widespread placental dysmorphology [13 , 14] . It has long been known that human and mouse placentas are hypomethylated compared to other tissues [15 , 16 , 17 , 18 , 19] . However , recent analysis of human placenta has shown a large-scale pattern of PMDs and HMDs that are often over 200 kb in length and can cover entire genes and gene clusters [20 , 21] . Human placenta PMDs cover tissue-specific genes that are transcriptionally repressed . These findings provide a unique opportunity to use DNA methylation to study not only the evolution of transcriptional regulation in this developmentally important organ but also the molecular similarities between species with morphologically distinct placentas . One study has shown that gene-specific methylation in mammalian placentas tracks with phylogeny more than placental morphology [22] , but otherwise little is known about the evolution of DNA methylation and transcriptional regulation in placenta . The placenta derives from the trophectoderm layer of the blastocyst in the early embryo , before the implantation stage . The DNA methylation patterns of the early embryo have recently been elucidated . Human and mouse oocytes have low levels of methylation which decrease even further in the fertilized embryo until after the blastocyst stage [23 , 24 , 25 , 26 , 27] . Therefore , it is thought that the placenta may never undergo the wave of remethylation that occurs in cells with other somatic tissue fates [28] . We performed MethylC-seq on a representative set of mammalian placentas , both with respect to evolutionary relationship and placental morphology . We found that although PMD/HMDs were not found in the placentas of many species , what was conserved was high methylation over gene bodies , particularly over those of active genes . Thus , in many mammalian placentas high methylation is found over single genes instead of over entire clusters of genes , as is the case in placentas with PMD/HMDs . Low genome-wide methylation and high gene body methylation in transcriptionally active genes was also found in the opossum EEM , suggesting that this is a conserved feature in the evolution of the mammalian placenta . Finally , analysis of DNA methylation in oocytes and preimplantation embryos shows that low global methylation but higher methylation over gene bodies is present prior to fertilization and persists through the blastocyst stage
Low coverage MethylC-seq was performed on rhesus , squirrel monkey , mouse , dog , cow , and horse placentas as well as opossum EEM ( S1 Table ) . Fig 1A shows the phylogenetic relationship between the species in this study and the type of placental morphology in each . In order to compare the pattern of DNA hypomethylation across mammalian placentas , average methylation was calculated for non-overlapping 20 kb windows tiled across the autosomes for each species . To compare placenta methylation levels to those of a somatic tissue , we also performed MethylC-seq on adult cow and dog cerebrum samples as well as fetal opossum whole brain . Fig 1B shows that mammalian placentas vary greatly in their overall amount of DNA methylation , although all were globally hypomethylated compared to brain , including the opossum EEM . All placentas had lower than 66% average methylation ( calculated as the average methylation of all CpG sites in all reads that aligned to the genome ) , but cow in particular had a remarkably low level of methylation in placenta ( 30% ) , even though cow brain global methylation is comparable to that of other mammals . In spite of this , methylation levels in regulatory elements such as CpG islands and promoters are similar between species ( S1 Fig ) and repetitive elements in general follow similar patterns , with LINEs and LTRs having methylation levels similar to those of non-repetitive regions and SINEs having relatively higher methylation ( S2 Fig ) . While human , squirrel monkey , and dog placentas show a bimodal distribution of methylation levels indicative of PMD and HMDs , other species showed a single peak or , like rhesus , had evidence of a small number of HMDs seen as a small secondary peak at higher methylation values . These results were similar for window lengths of 5–50 kb ( S3 Fig ) . Since the species showing PMD/HMD divisions were not necessarily the most closely related to humans , we next examined the conservation of methylation patterns at the chromosome and gene level . Homologous chromosomal regions of each species were mapped to the human genome using the liftOver program ( see Methods ) . Fig 1C shows that , despite the weak evidence for PMD/HMDs in the rhesus methylation density curves , methylation patterns across the rhesus chromosomes follow the patterns in human placenta remarkably well , with a Pearson correlation of 0 . 69 in 20 kb windows , the highest of all the species in this study ( S2 Table ) . Interestingly , despite the vast differences in the medians and interquartile ranges of window methylation levels between the different species , the correlations are remarkably high , with mouse placenta having the least similarity with human placenta of the species studied ( S2 Table and Figs 1C and S4 ) . Mapping a species genome to the human genome can introduce errors and artifacts , particularly in distantly-related species such as opossum ( S4 Fig ) . To eliminate these we next examined two loci with high methylation conservation between the species without mapping to the human genome ( Figs 1D , S5 and S6 ) . Figs 1D and S5 show the CNTNAP2 locus which includes two genes involved in neuronal development , CNTNAP2 and DPP6 , as well as the gene encoding Polycomb group protein EZH2 . Since neuron-specific genes tend to be in PMDs in human placenta [21] , we asked if hypomethylation was a conserved feature of this neuronal gene locus in the placenta . Figs 1D and S5 show that in most species both CNTNAP2 and DPP6 are in regions of relatively lower methylation compared to the surrounding locus . Most striking , however , are the short regions of relatively high methylation over individual genes or small clusters of genes that are highly conserved across all species , including the opossum EEM . EZH2 , KMT2C ( a histone methyltransferase expressed in placenta ) , PAXIP1 ( a gene that maintains genome stability during mitosis ) , and RBM33 ( a hypothetical RNA-binding protein ) have higher than average methylation in all the species studied . Likewise in the DAB1 locus ( S6 Fig ) , genes such as USP24 ( a ubiquitin peptidase ) , FGGY ( phosphorylates carbohydrates ) , INADL ( scaffolding protein in the cell membrane ) , DOCK7 ( a guanine nucleotide exchange factor ) , and ATG4A ( cysteine protease required for autophagy ) are highly methylated in most species . Our previous MethylC-seq analysis of three full-term human placentas showed that PMDs and HMDs were highly reproducible across individuals and gestational ages [21] . Here we performed five additional experiments to test the sensitivity of placenta methylome data to sampling differences and cellular heterogeneity . First , MethylC-seq analysis of whole rhesus placental tissue and isolated rhesus trophoblast cells gave nearly identical results with a correlation of 0 . 89 ( S7A , S7B , and S7C Fig ) . MethylC-seq results for E15 . 5 C57Bl6/J mouse placenta are very similar to those previously reported by Hon et al . [19] and a E11 . 5 C57BI6/J placenta sample ( S7D , S7E and S7F Fig ) with pairwise correlations over 0 . 87 . Since cow placenta has such low methylation , to validate the results we sequenced three additional biological replicates using tissue material from a different source and found nearly identical results with correlations above 0 . 75 ( S8A , S8B , and S8C Fig ) . We also sequenced two additional dog placentas from different breeds and developmental timepoints ( S1 Table ) . Methylation patterns across the chromosomes were very similar and pairwise correlations were over 0 . 87 ( S8D and S8F Fig ) . Although one of the dog placentas did not show a bimodal distribution whereas the other two did ( S8E Fig ) , the relative pattern of methylation along chromosomes was highly similar and the pairwise correlations were high ( 0 . 89 and 0 . 87 ) . Thus , the length of the range of methylation values in the curve may be more important than its bimodality . Finally , to confirm the reproducibility of human placenta MethylC-seq across samples , labs , and sequencing coverage , we compared our three previously sequenced placentas [21] to that of a higher-coverage MethylC-seq sample [29] . S9 Fig shows that all four human placenta samples have nearly identical global methylation patterns with pairwise correlations of >0 . 95 , demonstrating that sequencing depth does not affect global methylation levels or patterns . To determine the functions of genes with conserved high methylation in mammalian placentas , orthologous Ensembl genes in the seven species were clustered based on average gene body methylation ( Fig 2A ) . We found a group of 3380 genes ( branches A and B ) that had relatively high methylation within each species . This group is enriched for functions such as cell cycle , protein localization , and protein ubiquitination ( S3 Table ) . In contrast , genes with consistently low methylation across species tended to be transcription factors and have developmental functions ( branch G ) , of which 30% are polycomb-regulated . Our previous study found that human placenta also has low methylation over some tissue-specific genes such as those involved in neuronal functions of synaptic transmission and ion transport [21] . Across species , however , this trend of lower methylation of neuronal genes is not as conserved since dog , horse , and opossum show mixed or higher levels of methylation in these genes ( branch F ) . An interesting question is whether conservation of long-range methylation patterns ( eg . human and rhesus in Fig 1C ) is due to conservation of large chromatin domains or whether such patterns arise at the gene level . To address this question , we compared regions with large chromosomal breaks in synteny with regions of differential relative methylation between human and other species ( S10A Fig ) . Fig 2B shows a portion of human chromosome 7 with a relatively large number of chromosomal syntenic breaks between human and rhesus . Syntenic breaks are not enriched for differential methylation , suggesting that methylation patterns are established at a more local level . Since we had shown that DNA methylation was most conserved in gene bodies , we examined more closely the relationship between DNA methylation and genes . S10B Fig shows that the boundaries of human placenta PMDs are closer to gene ends and CpG islands than expected by chance . We next asked if DNA methylation is predictive of gene location in the placentas of all species studied . Non-overlapping 5 kb windows were tiled across the autosomes of each species and classified as genic or intergenic . S11 Fig shows that intergenic regions have lower DNA methylation levels than genic regions . Figs 2C and S12 show the probability that a window is in a gene based on its average % methylation . The further the probability is from 0 . 5 , the more informative the methylation level is in predicting the presence ( above ) or absence ( below ) of a gene . In all species studied , including opossum , high placental methylation was associated with genes , as was very low methylation since CpG islands were not removed . This is in contrast to brain where high methylation gives little to no information about the presence of a gene . Together these data indicate that DNA methylation patterns in placenta are being set at the gene level . We next asked whether , as in human placenta , genes with high methylation are more likely to be expressed across a broad range of mammals with diverse placental anatomies [21] . Published polyA-selected RNA-seq data for human , mouse , and horse placenta and opossum EEM [30 , 31] were utilized to compare gene expression to gene body methylation in orthologous genes . In human placenta , expressed genes have clearly higher than average gene body methylation levels ( Fig 3A , left column ) . To a lesser extent this can also be observed in mouse , horse , and opossum , although the range of gene body methylation values is smaller than human . To determine if these distributions are different than we would expect by chance given the marginal gene body methylation and gene expression distributions , we divided the x and y axes into 20 equally-spaced bins , counted the number of observations in each resulting quadrant , and compared that to the expected number of observations if gene body methylation and gene expression were independent ( Fig 3A , right column ) . A co-independence test showed a statistically significant deviation from independence for all four species , but more importantly the patterns of deviation are remarkably similar between the species . In all species examined , genes with high gene body methylation are more likely to have intermediate expression than expected by chance and genes with low methylation are less likely to be expressed . An example of the overlap between methylation and expression is the HECTD1 locus in Fig 3B . The HECTD1 gene encodes an E2 ubiquitin protein ligase that is important for proper placenta development in mouse [32] . HECTD1 has higher than average methylation levels and is expressed in human , mouse , and opossum . This is also true for the neighboring gene SCFD1 , which encodes a protein involved in SNARE-pin assembly and vesicular transport . An important question is whether mammalian placentas inherit their distinct methylation patterns from the early embryo or the patterns instead emerge later during placental development . To answer this question we utilized a combination of mouse oocyte and early embryo methylation data [24] and human oocyte data [27] and also performed MethylC-seq on MII cow oocytes . Similar to what was observed in mouse and human oocytes [23 , 27] , cow oocytes exhibit a bimodal methylation distribution ( Figs 4A , left column , and S13 ) , a surprising result since mouse and cow placentas do not show bimodal distributions . In addition , although both human oocytes and placenta have bimodal methylation distributions , the global patterns are quite dissimilar ( S14A Fig ) . We next looked at gene body methylation . A bimodal distribution of average gene body methylation was also observed in all three species' oocytes , although this is more prominent in the human and mouse oocytes ( Fig 4A , center column ) . In mouse , where MethylC-seq data is available for multiple early developmental timepoints , gene body methylation shows bimodal distributions that persist through the inner cell mass ( ICM ) stage ( S13 Fig ) and gene body methylation patterns remain correlated between the oocyte and ICM stages ( S15 Fig ) suggesting that although global methylation levels drop , relative levels of methylation in gene bodies remains consistent . Using oocyte expression data [24 , 33 , 34] we found that , similar to what was seen in placenta , methylation is enriched in gene bodies in oocytes and the mouse early embryo ( S16 Fig ) and genes with high gene body methylation are more likely to be expressed ( Fig 4A , right column ) . Genes with high methylation in the oocytes of all three species are enriched for functions related to vesicle-mediated transport , protein phosphorylation , and chromosome organization ( Fig 4B and S3 Table ) , similar to that seen in mammalian placentas ( Fig 2A and S3 Table ) . In fact , genes with high expression in both placenta and oocytes are enriched for these same functions ( S17A and S17B Fig ) and there is a large amount of overlap between the genes that are highly methylated in placenta and oocytes and those that are highly expressed in placenta and oocytes . Thus , these genes appear to maintain a similar regulatory pattern in the placenta that was set up very early in development .
In this first comparative study of genome-wide methylation patterns in mammalian placentas , we have made several novel findings of relevance to understanding the association of transcription and methylation in early life . First , we confirm that hypomethylation of extraembryonic tissues compared to somatic tissues is observed across eutherian mammals and a metatherian mammal as well . Second , we demonstrate that while large differences exist between mammalian placentas in global methylation levels , higher relative methylation of active genes is an evolutionary conserved feature . Third , we show that relative methylation is predictive of genic location in placenta . Lastly , we establish that the pattern of higher gene body methylation of active genes is also observed in mammalian oocytes and persists in the preimplantation embryo . While hypomethylation of placenta and EEM compared to somatic tissue was observed across mammalian species , global levels of placental methylation and the presence/absence of a bimodal PMD/HMD organization was diverse . The diversity of global methylation levels in mammalian placentas is not entirely surprising given the diversity of mammalian placenta morphologies . Based on current models of the mammalian radiation , it would appear that PMD/HMDs arose or were lost multiple times during placental evolution . However , our analysis of mouse and cow oocytes shows that even though some species do not have PMD/HMDs in their placentas , PMD/HMDs may still exist in the oocytes and early embryos , suggesting that these methylation patterns are later lost during placenta development in those species . One potential limitation of this study is the inherent diversity of placental morphologies across species and issues of cellular heterogeneity . Differences in absolute methylation levels and the existence of PMD/HMDs could have to do with these confounding factors rather than real differences between species . However , analysis of placental biological replicates in multiple species shows that while absolute global levels of methylation may vary up to 20% due to tissue sampling and inter-individual differences , the methylation patterns across a chromosome remain remarkably similar . Thus , our conclusion of relatively higher methylation over active genes as a conserved feature of mammalian placentas is unaffected . Another potential limitation is the relatively low MethylC-seq coverage used in this study , as low as 1X coverage for most of the placentas used in this study ( S1 Table ) . However , Ziller et al . experimentally determined coverage recommendations for whole-genome bisulfite sequencing , demonstrating that 1X coverage was sufficient for analysis of differentially methylated regions over 5 kb in length with methylation differences over 20% given two sequencing replicates [35] . One interesting question is why PMDs are hypomethylated . It was previously found in IMR90 fetal lung fibroblast cells that PMDs have a unique methylation distribution at individual CpG sites compared to HMDs [36] . High-coverage MethylC-seq data showed that within PMDs , each CpG site has a seemingly random level of methylation , but with some correlation in methylation between neighboring CpG sites less than 100 bp away . Moreover , the best predictor of the methylation level at a CpG site in PMDs was the distance to the nearest neighboring CpG site . In PMDs , unlike in CpG islands , CpG density is positively correlated with methylation . The authors hypothesize that PMDs may represent regions with reduced access to DNMTs and that in such an environment subtle DNMT sequence binding preferences may be easier to detect . Why do extraembryonic and placental tissues exhibit low global methylation and high gene body methylation , but somatic tissues such as brain are globally highly methylated ? One hypothesis stems from the observation that hypomethylation and well-defined PMD/HMD structures are also found in cancerous tumors [37 , 38 , 39] . Remarkably , the presence of hypomethylated PMDs is the most epigenetically defining feature of a wide range of human solid tumors [40] , EBV transformation of B cells is characterized by the gain of hypomethylated PMDs [41] , and brain tumors gain hypomethylated PMDs as they progress to higher metastases [42] . Since placenta shares many features in common with metastatic tumors , including rapid proliferation , invasiveness , and angiogenesis [43 , 44 , 45] , perhaps a hypomethylated state would be evolutionary advantageous in a temporary tissue in which growth is highly regulated by pregnancy hormones . The temporary nature of the placenta may also make is less susceptible to the effects of retrotransposable elements . In contrast , somatic tissues in more long-lived mammalian species may have evolved globally higher methylation levels as a mechanism to prevent cancer in other tissues .
The sources of the placenta and brain samples sequenced for this study are listed in S1 Table . Human subjects were approved by the UC Davis IRB ( 225645–17 ) and maternal written consent was obtained . All procedures involving animals were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals and under the approval of the University of California Davis , Animal Care and Use Committee ( Animal Protocol #15639 ) . Where possible , villus tissue was taken from the fetal side or interior ( equidistant from the fetal and maternal sides ) of the placenta , excluding large blood vessels , membranes , and any obvious connective tissue . Opossum EEM tissue was collected as described [46] . Samples were stored frozen at -80 degrees Celsius . Based on similarly processed and stored samples of human fetal side placenta , maternal cell contamination is expected to contribute <10% to methylation and sample location and inter-individual differences are expected to contribute to up to 20% differences in percent methylation over PMDs [21] . Trophoblast cells were isolated from Rhesus monkey ( Macaca mulatta ) placental tissue ( Gestation Day: 40–65 days ) using procedures we have previously described [47 , 48 , 49] . All procedures involving animals were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals and under the approval of the University of California Davis , Animal Care and Use Committee ( Animal Protocol #15639 ) . These cells are 95% cytokeratin 7-positive and 5% vimentin-positive , consistent with a predominantly trophoblast population . MII oocytes were produced by in vitro maturation of GV oocytes collected from slaughter house derived ovaries according to standard protocols . Oocyte maturation was confirmed by presence of a polar body . Then , zona pellucida and first polar body were removed by incubation in 0 . 5% pronase solution for 2 minutes and vigorous pipetting . Zona-free oocytes were snap frozen in liquid nitrogen and stored at -80°C until DNA extraction . DNA from rhesus , cow , horse , dog , squirrel monkey , and mouse placenta tissue was purified using Qiagen's Puregene kit . MethylC-seq libraries were made as described previously [21] . Briefly , the genomic DNA was sonicated to ~300 bp and methylated Illumina adapters were ligated to the ends . The library was bisulfite converted , amplified for 14 cycles , and sequenced on either an Illumina HiSeq or GAII . For the opossum EEM , rhesus trophoblast , human cord blood , and cow , dog , and opossum brain samples , MethylC-seq libraries were made using the Epicentre EpiGnome Methyl-Seq kit according to the manufacturer's recommendations except that 14 cycles of amplification were performed . For cow oocytes , DNA from 100 cells was purified using Zymo's Quick-gDNA MicroPrep kit and the MethylC-seq libraries were prepared using Zymo's Pico Methyl-seq Library Prep Kit according to the manufacturer's instructions . After sequencing on HiSeq2500/2000 machines , reads were mapped to the respective genomes using BS Seeker [50] and only one read per genomic position was kept to prevent clonal PCR amplification biases . CpG site methylation data were combined from both DNA strands . Because methylation data was analyzed over large genomic distances and/or smoothed , no minimum coverage was required for CpG sites used in this analysis [35] . Individual CpGs were mapped to coordinates on the human genome using the liftOver program and species-specific liftOver chain files available on the UCSC Genome Browser [51] . Since our analysis focused on global methylation patterns , not individual CpG sites , mappings were done regardless of whether the CpG site was conserved in human . Since our analysis was focused on methylation in large syntenic regions , we removed CpG data in small inter- and intrachromosomal translocations , duplications , etc . using the species-specific synteny net files available on the UCSC Genome Browser . CpGs were removed if they were in the second level fill of the synteny net files and were less than 1 Mb in length . For graphing the cross-species global methylation patterns , these “cleaned” species liftOver data and raw human data were compressed into averages of non-overlapping 20 kb windows ( windows with less than 20 CpG sites with methylation information were discarded ) and smoothed in R using a kernel smoother . For graphing the cross-species syntenic breaks , only the breaks between the large syntenic regions in the first level of the UCSC synteny net files and those between large ( >1 Mb ) syntenic regions in the second level were used since smaller non-syntenic regions had been removed from the methylation datasets . Due to the fact that the squirrel monkey genome ( saiBol1 ) was not yet assembled into chromosomes , many of the “syntenic breaks” between human and squirrel monkey are actually between contigs that may or may not be on the same squirrel monkey chromosome . After mapping species methylation data to the human genome , the data was smoothed and normalized . Average methylation was taken for non-overlapping 20 kb windows and a running median was computed using a width of 15 windows . The data was then scaled so that each species had the same mean and standard deviation . The smoothed and scaled methylation values were subtracted from the smoothed and scaled human methylation values and regions of differential methylation were defined as those with methylation differences over 1 . 5 standard deviations . Gene annotations for each species as well orthologous gene information were obtained from Ensembl's biomart . Promoter and CpG island sequences were removed from each gene before calculating average percent methylation over gene bodies . Genes with fewer than 20 remaining CpG sites with methylation information were removed from each species' dataset . | The placenta is vital for the proper development of the fetus , not only facilitating the exchange of nutrients , oxygen , and waste between the mother and the fetus but also acting as an interface to the maternal immune system and regulating fetal growth by excreting hormones and growth factors . DNA methylation is important for both placental and embryonic development as loss of proteins involved in DNA methylation can result in placental dysmorphology and early embryonic death . The human placenta has a unique DNA methylation landscape characterized by alternating regions of low methylation , covering silent genes with tissue-specific developmental functions , and high methylation , covering active genes . In order to better understand the significance of this DNA methylation landscape in the human placenta , we performed a cross-species comparison of DNA methylation in mammalian placentas , oocytes , and early embryos from this and other studies . Although the levels and extent of hypomethylation differed between mammalian placentas , what we found to be highly conserved was relatively higher methylation levels over active genes . These same genes also had high methylation in the opossum extraembryonic membrane , a primitive placenta , as well as oocytes and early embryos , suggesting that high methylation over these genes predated placental mammals and is established very early in development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Early Developmental and Evolutionary Origins of Gene Body DNA Methylation Patterns in Mammalian Placentas |
What role does attention play in ensuring the temporal precision of visual perception ? Behavioural studies have investigated feature selection and binding in time using fleeting sequences of stimuli in the Rapid Serial Visual Presentation ( RSVP ) paradigm , and found that temporal accuracy is reduced when attentional control is diminished . To reduce the efficacy of attentional deployment , these studies have employed the Attentional Blink ( AB ) phenomenon . In this article , we use electroencephalography ( EEG ) to directly investigate the temporal dynamics of conscious perception . Specifically , employing a combination of experimental analysis and neural network modelling , we test the hypothesis that the availability of attention reduces temporal jitter in the latency between a target's visual onset and its consolidation into working memory . We perform time-frequency analysis on data from an AB study to compare the EEG trials underlying the P3 ERPs ( Event-related Potential ) evoked by targets seen outside vs . inside the AB time window . We find visual differences in phase-sorted ERPimages and statistical differences in the variance of the P3 phase distributions . These results argue for increased variation in the latency of conscious perception during the AB . This experimental analysis is complemented by a theoretical exploration of temporal attention and target processing . Using activation traces from the Neural-ST2 model , we generate virtual ERPs and virtual ERPimages . These are compared to their human counterparts to propose an explanation of how target consolidation in the context of the AB influences the temporal variability of selective attention . The AB provides us with a suitable phenomenon with which to investigate the interplay between attention and perception . The combination of experimental and theoretical elucidation in this article contributes to converging evidence for the notion that the AB reflects a reduction in the temporal acuity of selective attention and the timeliness of perception .
An apparent temporal limitation of visual perception is illustrated by the attentional blink ( AB; [6] ) . The AB describes a finding that observers often fail to detect a second target stimulus ( T2 ) presented in short succession ( between 100 and 600 ms ) after an identified first target stimulus ( T1 ) . If T2 is presented in immediate succession to T1 , however , detection accuracy is typically excellent ( ‘lag 1 sparing’; [7] ) . Behaviourally , the AB has been replicated numerous times [8] , [9] . It has also been investigated electrophysiologically [10] , where researchers have compared grand average Event-related Potentials ( ERPs ) evoked by targets outside and inside the AB , to investigate how target processing differs during the AB . Despite extensive study of the AB , its effect on the underlying temporal mechanisms of target identification remains to be fully explored . Evidence from ERP [10] , [11] and priming [12] , [13] studies suggest that targets , rather than being completely lost during the AB , are processed quite extensively , but fail to enter the final stage of conscious perception . Furthermore , researchers have found that when targets in RSVP consist of multiple features , observers often report features from items neighboring the target in the RSVP stream and make binding errors referred to as illusory conjunctions [14] . Behavioural analysis of the changes in the patterns of such binding errors provides strong support for the claim that the AB reveals a reduction in the temporal precision of the deployment of transient attention and target processing [15] , [16] . In this article , we use the dynamics of temporal visual processing as embodied in the ( Simultaneous-Type-Serial-Token ) model , a connectionist model of temporal attention and working memory [5] , to propose an explanation for the observed effect of the AB on the temporal precision of transient attention . The model explains a broad set of experimental findings relating to the AB , Repetition Blindness and RSVP in general . Before elaborating on our central hypothesis , we explain the fundamental principles of how the model describes temporal attention and working memory . For a more detailed description please refer to [5] . It should be emphasised that throughout this article , we retain the model's parameters as published in [17] , and use it to generate predictions and virtual EEG traces comparable to human EEG data . The model suggests that working memory encoding involves creating a binding between the type of a stimulus ( which can include its visual features and semantic attributes ) and a token ( an episodic representation specific to a particular occurrence of an item ) [18] , . In the model , Transient Attentional Enhancement ( TAE ) from the blaster amplifies the type representation of a salient ( i . e . , task relevant ) stimulus to assist in its binding to a token , in a process referred to as tokenization . This TAE can serve as an attentional gate , which can be temporarily deactivated to allow one target's encoding to be completed before a second is begun . From the perspective of the model , the AB is an artifact of the visual system attempting to assign unique tokens to targets [22] . More specifically , the process of encoding T1 into working memory is triggered by TAE , and TAE itself is subsequently suppressed until T1 encoding has completed . The period of TAE unavailability varies from trial to trial depending on how long it takes to tokenise T1 , depending on its bottom-up strength . In an RSVP stream , if a T2 is presented 100–600 ms after a perceived T1 ( as is the case during the AB ) , its processing outcome depends on multiple factors . T2's own strength determines its dependence on TAE , since highly salient T2s can ‘break-through’ the AB [31] and get encoded relatively early . T2s with strength values slightly lower in the range ‘outlive’ the AB ( and thus the unavailability of TAE ) , and hence are indirectly influenced by T1 strength . Overall , the variability in the temporal dynamics of T2's encoding process is influenced both by T1 and T2 strengths . Hence , over all possible strengths , the model proposes that there should be increased variance in processing latency for targets seen during the AB . This article investigates the hypothesis that diminished attentional control increases the temporal jitter in the latency of a target's working memory consolidation . The AB provides us with a suitable phenomenon with which to test our hypothesis: we propose that the reduced availability of attention during the AB increases the temporal noise in visual attention . To answer this question , we compare the EEG signatures evoked by targets seen outside vs . inside the AB , and determine whether there is a comparative increase in the variability of the latency of working memory encoding of targets presented inside the AB . EEG has the advantage of excellent temporal resolution , allowing us to study short-lived cognitive events that evoke changes in ongoing EEG activity . If one averages over multiple segments of such EEG activity time-locked to the event , the resulting averaged ERP waveform contains a number of positive and negative deflections , referred to as ERP components . To test for increased temporal jitter , we analyse the P3 ERP component , commonly associated with encoding items into working memory [10] , [23] . However , analysis of averaged ERP components cannot directly inform our hypothesis . This is because the averaging collapses across and hence discards information about temporal fluctuations in the individual EEG trials contributing to the ERP . Given a set of trials that are averaged together , both decreases in amplitude and increases in latency variation within that set will attenuate the mean amplitude of the ERP . Hence , examining the average does not directly provide the necessary information to decide which of the two sources of variation in the individual trials ( amplitude or latency ) caused the reduction in ERP amplitude . Further , measures like 50% area latency analysis [24] cannot be used to measure latencies in single trials , due to the levels of irrelevant noise activity . Consequently , we employ time-frequency analysis techniques that provide alternative measures to investigate single trial dynamics underlying grand average ERPs . These methods enable us to perform a more fine-grained analysis of EEG data , and test our hypothesis using both qualitative and quantitative means . In addition to presenting and analyzing human EEG data , we use the model's neural network implementation to generate virtual P3 ERP components [17] , which are hypothesised to correspond to the human P3 ERP component . For each of the experimental conditions , the virtual P3 is contrasted with the human P3 , both at grand average and single trial level . This comparative evaluation allows us to validate the model and propose explanations for the human ERP effects .
The experiment consisted of RSVP trials presented at a rate of 105 . 9 ms per item , with two letter targets , T1 and T2 , embedded among digit distractors . T2 was presented at lags 1 , 3 and 8 following the T1 . The P3 EEG data analysed in this section was recorded at the Pz electrode . Please refer to the Materials and Methods section for further information . Mean human accuracy for T1 identification was 82% . The accuracy of T2 identification ( conditional on correct report of T1 ) was 83% at lag 1 , 54% at lag 3 , and 74% at lag 8 . There was a significant effect of lag on accuracy ( F ( 1 . 48 , 12 . 58 ) = 15 . 58 , MSE = 0 . 03 , p0 . 001 , after applying a Greenhouse-Geisser correction on the degrees of freedom ) . Additionally , in pairwise comparisons , T2 accuracy was significantly lower at lag 3 compared to lag 8 ( F ( 1 , 17 ) = 11 . 66 , MSE = . 03 , p = . 003 ) and lag 1 ( F ( 1 , 17 ) = 60 . 88 , MSE = 0 . 01 , p0 . 001 ) . Consequently , the paradigm employed in this study evoked a reliable AB effect . The ERPimages [25] in figure 2 compare the P3 evoked by targets seen outside the AB ( seen T2s at lag 8 following a seen T1 ) with targets seen inside the AB ( seen T2s at lag 3 following a seen T1 ) . They allowed us to visualise the EEG trials underlying the grand average P3 ERPs ( plotted below them ) for targets seen outside and inside the AB . These ERPimages represent time with respect to target onset along the X-axis ( Note that trials are time-locked to T2 onset ) , individual trials along the Y-axis , and the single-trial EEG amplitude using a colour scale . The trials comprising these images were sorted from bottom to top by descending order of the phase angle of the single-trial P3 at the time point indicated by the dashed line , which was set to the peak latency of the corresponding grand average P3 . This phase angle was estimated at the frequency at which the power of the P3 was maximal . This sorting method effectively ordered the trials according to the approximate latency of the single-trial P3 for a target , as estimated by a wavelet-based time-frequency analysis ( see the Materials and Methods section for more details ) . The ERPimages were then plotted for each condition , with trials having longer latency P3s being placed at the bottom , and trials with shorter latency P3s at the top . Following from our hypothesis , for targets inside the AB , we expected to observe an increased “slope” in the red “smear” representing the P3 . This would indicate that these targets suffer greater temporal variance compared to targets outside the AB . A visual comparison of the ERPimages clearly suggested that the P3 for targets outside the AB ( figure 2A ) had relatively little variation in the phase angle across most trials . In other words , the P3 onset occurred at approximately the same time in these trials . In contrast , the P3 evoked by targets inside the AB ( figure 2B ) appeared to exhibit an increased temporal fluctuation , as reflected by the increased variance of the phase angle of the P3 across all trials . A natural consequence of this jitter in the temporal onset of the P3 was a ‘smearing out’ of the grand average ERP . In summary , if there was indeed a reduction in the precision of the deployment of attention in response to targets during the AB , we expected this to indirectly affect the working memory encoding of targets as reflected by the P3 . The ERPimages in figure 2 provided qualitative support for our hypothesis . We then extended this investigation by analysing the distribution of phase angles corresponding to the P3 , to generate numerical evidence that could be verified statistically . To back up the qualitative comparisons of the previous section , we statistically analysed the time-frequency data obtained therein . We used an approach similar to inter-trial phase coherence analysis [25] , but adapted the idea to directly examine the subject-wise P3 phase distributions and quantitatively compare temporal jitter . The phase angles used to sort the individual trials comprising the P3 ERPimages in the previous section formed a circular distribution [26] of angular data values that effectively represented the temporal latency between the onset of the target and its P3 . By statistically comparing the variance in the distribution of phase angles across targets outside and inside the AB , we tested whether the visual differences observed were consistent across subjects . To do so , we performed a subject-wise grouping of the P3 phase angles calculated at the peak latency of the grand average P3 for each condition ( the same phase angles that were used to sort the ERPimages presented earlier ) . This generated multiple smaller distributions of P3 phase angles , one per condition and subject . These distributions were then modelled as von Mises distributions [26] for which the concentration parameter was calculated using maximum likelihood estimation . The parameter of a distribution is a measure of its density around its mean value , and is an analogue of the inverse of its variance . The larger the value of a circular distribution , the more concentrated it is around the mean . Importantly , is a linear parameter , and can be compared using conventional statistical tools . Hence , in order to test whether targets inside the AB suffered from increased temporal jitter , we compared values of the subject-wise P3 phase distributions evoked by targets outside and inside the AB , using a standard one-way repeated-measures ANOVA . The results of the ANOVA validated what the visual differences observed in the ERPimages clearly indicated: The of the phase distribution for the P3 for targets outside the AB was statistically greater than that for targets inside the AB: Mean for targets outside the AB was 0 . 95 , whereas mean for targets inside the AB was 0 . 52 ( F ( 1 , 17 ) = 15 . 21 , MSE = 0 . 11 , p = 0 . 001 ) . In order to validate the model , we used it to generate ‘artificial electrophysiological’ traces , so-called virtual ERPs [17] . In analogy to human ERP components , we generated virtual ERP components for targets outside and inside the AB . This approach , in addition to allowing us to validate the internal dynamics of the model , provided theoretical explanations for the human EEG effects observed in the previous section . Please refer to the Materials and Methods section for more details on how virtual ERPs and ERPimages were generated .
Our experimental results and theoretical explorations complement and inform previous research on temporal selection and the AB . We now discuss these findings and propose interpretations in terms of the model . In this article , we have presented human ERP evidence in favour of a reduction in the temporal precision of transient attention during the AB . The AB provides us with a suitable phenomenon with which to investigate the interplay between attention and perception . The interplay between these tightly linked cognitive processes is adversely affected during the AB , producing the reduction in precision observed in behavioural and EEG data . Using ERPimages , we have provided qualitative evidence arguing for an increase in temporal variation in the dynamics of P3s evoked by targets seen outside vs . inside the AB window . This evidence is supported quantitatively , by statistical comparison of the phase distributions corresponding to the P3 . This analysis suggests that there is significantly increased temporal jitter in the ERP activity evoked by targets inside the AB . This notion of a decrease in the temporal precision of attention is inherent in the theoretical framework of the model . Specifically , we have used the model’s neural implementation to generate both virtual ERPs and ERPimages , which we have then compared to their human counterparts . We believe that correlating model and electrophysiological data in this way provides a two-fold benefit . Firstly , it has provided a sufficient explanation for the modulatory effects of the AB on the temporal precision of visual processing . Secondly , it has allowed us to instantiate and test the model at the level of single-trial dynamics , and show that the theoretical assumptions about the nature of temporal visual processing embodied by it can be validated using EEG data , in addition to traditional behavioural verification . We believe that the combination of experimental and theoretical analysis presented in this article contributes to converging evidence for the notion that the AB results in a reduction in the temporal acuity of selective attention , which is an important mechanism for ensuring the timeliness of conscious perception .
This section describes the experiment ( the same as Experiment 2 from [17] ) used to collect the human EEG data analysed in this article . | Our visual system keeps pace with a rapidly changing stream of information as we view the natural world . To do so , it uses a strongly regulated system of attentional filters to constrain which visual stimuli are permitted to be fully processed to the level of conscious awareness . This article explores what happens when these filters are opened and closed in response to important visual stimuli . To understand these dynamics , our neural network model provides simulations of the role played by attention . These simulations can be tested by recording neural data in the form of ‘brain waves’ ( EEG ) and comparing the resultant signals to the output of the model . The data discussed here confirm a prediction of the model , which suggests that after the attentional filter has opened to allow one visual stimulus in , there is increased temporal variability or ‘jitter’ in the subsequent opening of the filter within an interval of about one-half of a second . These results have implications for the way our brains process multiple important stimuli perceived in rapid succession , such as the sequence of events that might occur at a critical moment in an airline cockpit or during an automobile accident . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/cognitive",
"neuroscience",
"neuroscience/theoretical",
"neuroscience"
] | 2009 | Attention Increases the Temporal Precision of Conscious Perception: Verifying the Neural-ST2 Model |
Sporothrix schenckii and associated species are agents of human and animal sporotrichosis that cause large sapronoses and zoonoses worldwide . Epidemiological surveillance has highlighted an overwhelming occurrence of the highly pathogenic fungus Sporothrix brasiliensis during feline outbreaks , leading to massive transmissions to humans . Early diagnosis of feline sporotrichosis by demonstrating the presence of a surrogate marker of infection can have a key role for selecting appropriate disease control measures and minimizing zoonotic transmission to humans . We explored the presence and diversity of serum antibodies ( IgG ) specific against Sporothrix antigens in cats with sporotrichosis and evaluated the utility of these antibodies for serodiagnosis . Antigen profiling included protein extracts from the closest known relatives S . brasiliensis and S . schenckii . Enzyme-linked immunosorbent assays and immunoblotting enabled us to characterize the major antigens of feline sporotrichosis from sera from cats with sporotrichosis ( n = 49 ) , healthy cats ( n = 19 ) , and cats with other diseases ( n = 20 ) . Enzyme-linked immunosorbent assay-based quantitation of anti-Sporothrix IgG exhibited high sensitivity and specificity in cats with sporotrichosis ( area under the curve , 1 . 0; 95% confidence interval , 0 . 94–1; P<0 . 0001 ) versus controls . The two sets of Sporothrix antigens were remarkably cross-reactive , supporting the hypothesis that antigenic epitopes may be conserved among closely related agents . One-dimensional immunoblotting indicated that 3-carboxymuconate cyclase ( a 60-kDa protein in S . brasiliensis and a 70-kDa protein in S . schenckii ) is the immunodominant antigen in feline sporotrichosis . Two-dimensional immunoblotting revealed six IgG-reactive isoforms of gp60 in the S . brasiliensis proteome , similar to the humoral response found in human sporotrichosis . A convergent IgG-response in various hosts ( mice , cats , and humans ) has important implications for our understanding of the coevolution of Sporothrix and its warm-blooded hosts . We propose that 3-carboxymuconate cyclase has potential for the serological diagnosis of sporotrichosis and as target for the development of an effective multi-species vaccine against sporotrichosis in animals and humans .
Sporothrix schenckii was originally described in 1898 as the causal agent of a subcutaneous disease in humans in the Mid-Atlantic USA [1] . Subsequently , the disease was reported in rats naturally infected in southeastern Brazil [2] and later in a wide range of animals including dogs , cats , horses , cows , camels , dolphins , goats , mules , birds , pigs , and armadillos . Several Sporothrix spp . , previously reported to be closely related to S . schenckii , are known to establish infections in various hosts , with dissimilar virulence traits [3 , 4] and responsiveness to antifungal treatment [5] . The S . schenckii complex consists of at least four closely-related species [6 , 7] , ranging from geographically restricted agents such as S . brasiliensis [8 , 9] to cosmopolitan pathogens such as S . schenckii s . str . and S . globosa [7 , 10 , 11] . Sporothrix spp . are endowed with an extraordinary ecological diversity [12–15]; they are frequently recovered from soil , plants debris , and insects ( Coleoptera: Scolytidae ) . Phylogenetic data support a recent habitat shift within Sporothrix from plants to cats [9] that culminated in the largest epizootic transmission in southeastern Brazil [16–19] . Feline sporotrichosis emerged in the 1990s , with S . brasiliensis recovered from many outbreaks [8 , 20] . More recently , S . brasiliensis has been recognized as a threat to humans [21–23] due to the massive zoonotic transmission in southeastern Brazil that affects thousands of patients regardless of whether they are immunocompetent or immunocompromised [9 , 24–26] . Cats have been a source of Sporothrix spp . infection transmitted to humans and other animals [18 , 19 , 27] . Most human cases occurred in housewives and professionals who had contact with infected animals and a history of scratches or bites [21 , 28] . The largest epidemic of sporotrichosis due to zoonotic transmission was reported in the State of Rio de Janeiro , Brazil [18 , 19 , 21 , 23 , 28]; since then , the incidence of sporotrichosis among animals , particularly cats , has increased [8 , 28 , 29] . More than 4 , 000 humans and 4 , 124 cats were diagnosed at Instituto Nacional de Infectologia ( INI ) Evandro Chagas /Fundação Oswaldo Cruz by 2012 [30] . Pereira et al . [29] observed that the majority of cats with sporotrichosis in Rio de Janeiro between 2005 and 2011 were male , mongrel , and unneutered , had a median age of 24 months , and presented with three or more cutaneous lesions in non-adjacent locations . This mycosis in cats is hard to treat and generally requires a long period of daily care; these animals do not always respond well to antifungal treatment [30] . Sporothrix is widely distributed in nature , and traumatic inoculation of plant organic matter is classically associated with infection [31] . Feline habits render cats more susceptible to the fungal agent because they are constantly in contact with soil , plant debris , and other cats that may be infected . In cats , a broad spectrum of clinical presentation ranges from single lesions to fatal systemic forms . After monitoring the feline epidemic for 4 years , Schubach et al . [18] observed that the lymphocutaneous form occurred less frequently than did involvement of the mucous membranes of the respiratory tract and upper digestive tract and multiple cutaneous lesions . Some animals present involvement of skin and various internal organs [32] . Cats are susceptible to a variety of fungal infections , including blastomycosis [33] , histoplasmosis [34] , cryptococcosis [35] , candidiasis [36] , dermatophytosis [37] , aspergillosis [38] , coccidioidomycosis [39] , and sporotrichosis . Misdiagnosis results in ineffective treatment , which further worsens outcome . Major contributors toward misdiagnosis include the small number of affordable and effective treatment techniques as well as other social issues . Definitive diagnosis of feline sporotrichosis is based on mycological culture , micromorphological characterization , and mold-to-yeast conversion . Histopathological methods and cytopathological examination are useful tools for the presumptive diagnosis of Sporothrix infection in cats [40] . Detection of specific anti-Sporothrix antibodies offers a rapid alternative for accurate diagnosis [41–43] . We recently proposed a serological approach that employs an enzyme-linked immunosorbent assay ( ELISA ) to diagnose feline sporotrichosis [44] using purified antigen ( the S . schenckii ConA binding fraction ) and crude antigen , with high sensitivity and specificity . There is a lack of information about feline sporotrichosis and the antigenic components involved in infection; therefore , the present study aimed to explore the diversity of molecules expressed by closely related species ( S . brasiliensis and S . schenckii ) and that are recognized by immunoglobulin G ( IgG ) in sera from cats naturally infected with S . brasiliensis . We found remarkable cross-reactivity among S . brasiliensis and S . schenckii antigens , and we identified an immunodominant molecule that is an important biomarker in feline sporotrichosis , irrespective of clinical manifestation . Here , we show that , although S . brasiliensis and S . schenckii may infect different warm-blooded hosts , infection result in highly similar IgG-mediated response in cats compared to humans , what is important for serodiagnosis and to the development of prophylactic or therapeutic vaccine against the enormous health burden of sporotrichosis in endemic areas . This knowledge may enable selection of potential biomarkers that can be used in seroepidemiological studies , diagnosis , and vaccine development , and may contribute to understanding of the pathogenesis of this infection in cats and humans .
This study was performed in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Ethics in Research Committee of the Fundação Oswaldo Cruz , Rio de Janeiro , Brazil , under license number L-041/06 . This study was also approved by the Institutional Ethics in Research Committee of the Federal University of São Paulo under protocol number 0244/11 . S . brasiliensis and S . schenckii s . str . from cats and humans were used for protein extraction ( Table 1 ) . The dimorphic nature of Sporothrix spp . was demonstrated by converting the fungus to its yeast form at 36°C in brain-heart infusion medium ( Difco ) and observing typical oval multibudding yeast cells . Molecular identification was performed and confirmed via DNA sequencing of the gene encoding calmodulin [25] . Isolates were selected because they had been previously characterized at the molecular level [3 , 9 , 45 , 46]; crude exoantigen ( CBS 132974 = Ss118; S . schenckii s . str . ) was successfully used to diagnose feline sporotrichosis via ELISA [44] . Sporothrix spp . was grown on Sabouraud medium ( Difco ) agar slants at room temperature ( 20–25°C ) for 7 days . Approximately 2x106 conidia ( ≥85% viable cells ) were used to inoculate 500-mL flasks containing 150 mL of brain-heart infusion broth . Cultures were incubated at 36°C in a rotary shaker ( Multitron II , Infors HT ) with constant orbital agitation at 110 rpm for 7 days . Whole extracts of Sporothrix yeast cells were obtained as described elsewhere [47] . Briefly , yeast cells ( 4 mL of each culture ) were collected via centrifugation at 5000 x g for 10 min at 4°C and washed three times in ultrapure water . Pellets were frozen in liquid nitrogen and disrupted by gridding . Cells were macerated with a pestle until a fine powder was obtained . This cellular powder was vortexed for 30 min at 4°C in Tris-Ca2+ buffer ( 20 mM Tris-HCl pH 8 . 8 , 2 mM CaCl2 ) containing a commercial cocktail of protease inhibitors ( 1:1000; GE Healthcare ) , RNAse and DNAse ( 1:1000; GE Healthcare ) , and 600-μm glass beads ( 1:1; Sigma ) . Cell debris and glass beads were removed via centrifugation at 5000 x g for 10 min , and dithiothreitol ( final concentration 20 mM ) was added to the supernatant [48] . Protein concentrations were determined using the Bradford method [49] . Cell extracts were kept at -80°C until use . Sera from cats with definitive diagnoses of sporotrichosis ( via S . brasiliensis isolation in culture; n = 49 ) were obtained from INI/Fundação Oswaldo Cruz , Rio de Janeiro , Brazil . The distribution and number of skin lesions of the cats were classified as L1 ( cutaneous lesion in only one place ) , L2 ( cutaneous lesion in two non-adjacent places ) , or L3 ( cutaneous lesions in three or more non-adjacent places ) . During examination , blood was collected via vein puncture and stored in an incubator for 1 h; serum was obtained via centrifugation and stored at -20°C until use . Sera from 19 cats with no evidence of sporotrichosis or other diseases ( the control group ) were obtained from São Paulo as described elsewhere [44] . Sera from 20 cats with other diseases were also studied to verify cross-reactions with feline infectious peritonitis/coronavirus ( 5 sera ) , feline leukemia virus ( 3 sera ) , feline immunodeficiency virus ( 2 sera ) , leptospirosis ( 3 sera ) , rickettsiosis ( 2 sera ) , erlichiosis ( 3 sera ) , and leishmaniasis ( 2 sera ) as previously described [44] ( S1 Diagram ) . All sera were stored at -20°C until use . Sera from cats with confirmed sporotrichosis and from cats from the control group were tested via ELISA . To determine the best protein concentration for microplate sensitization , whole cellular proteins from S . brasiliensis ( CBS 132990 and CBS 132021 ) and S . schenckii s . str . ( CBS 132974 and CBS 132984 ) were tested and examined by checkerboard titration for antibody detection . Afterward , all microplates were sensitized with concentrations of 3 . 6μg/mL ( 100μL per well in 0 . 1 M carbonate-bicarbonate buffer , pH 9 . 6 ) . High binding microtiter plates ( Corning Costar , Corning ) were sensitized for 2 h at 37°C and overnight at 4°C in a refrigerator . The remaining binding sites were blocked with phosphate-buffered saline containing 0 . 1% Tween 20 ( PBST ) and 5% non-fat dry milk ( 200 μL/well ) for 4 h at 37°C . After washing three times with PBST , diluted serum ( 1:800 in PBST , 100μL/well ) was added in duplicate for 1 h at 37°C . Afterward , 100μL horseradish peroxidase-conjugated goat anti-feline IgG ( 1:1000; Southern Biotech ) were added to each well and incubated for 1 h at 37°C . After three washes with PBST , 100μL substrate solution ( 5 mg of o-phenylenediamine in 25 mL of 0 . 1 M citrate-phosphate buffer pH 5 . 0 plus 10 μL 30% H2O2 ) were added to each well , and the reaction was interrupted after 8 min in the dark by adding 50 μL 4 N H2SO4 . Optical density was read at 492 nm with a Tecan Sunrise 96-well Microplate Reader ( Tecan ) . S . brasiliensis and S . schenckii protein extracts ( 2 μg ) were analyzed via SDS-PAGE with 10% gels [50] and silver-stained [51] . The relative molecular weights of the fractions were estimated using standard broad-range molecular weight markers ( Protein Benchmark , Invitrogen ) . For immunoblotting , proteins ( 10 μg ) from strains CBS 132990 , CBS 132021 , CBS 132974 , and CBS 132984 were resolved with SDS-PAGE and transferred onto 0 . 45-μm polyvinylidenedifluoride membranes ( Bio-Rad ) at 20 V for 30 min with transfer buffer ( 25 mM Tris base , 192 mM glycine , 20% methanol , pH 8 . 3 ) [52] using a Trans-Blot SD semi-dry device ( Bio-Rad ) . Electrotransference was confirmed by staining with 0 . 15% Ponceau S and 1% acetic acid [vol/vol] . Membranes were destained and free binding sites were blocked overnight in phosphate-buffered saline blocking buffer ( 1% bovine serum albumin supplemented with 0 . 05% [vol/vol] Tween 20 , 5% [wt/vol] skim milk , pH 7 . 6 ) at 4°C . To determine the best dilution of serum , one sample was tested at four dilutions ( 1:100 , 1:200 , 1:500 , and 1:1000 ) against yeast extracts . Afterward , for all sera , membranes were probed individually with primary antibody diluted 1:500 at 25°C for 2 h . Membranes were washed three times with Tris-buffered saline ( pH 7 . 5 ) containing 0 . 05% [vol/vol] Tween-20 for 10 min and incubated with horseradish peroxidase-conjugated goat anti-feline IgG ( 1:1000 ) for 2 h at room temperature . Membranes were then washed with Tris-buffered saline ( pH 7 . 5 ) containing 0 . 05% [vol/vol] Tween-20 and signal was detected with an enhanced chemiluminescence detection kit ( GE Healthcare ) . Blots were imaged in a transilluminator ( Uvitec Cambridge ) . Allience 4 . 7 software was used to take several images at different time exposures , from 2 s each to a total of 10 images over 2 s . Proteins were separated via 2D gel electrophoresis as previously described [45 , 47] . Briefly , proteins ( 300 μg ) were precipitated using the 2D Clean-up Kit ( GE Healthcare ) and resuspended in rehydration buffer ( 7 M urea , 2 M thiourea , 2% CHAPS , 1 . 2% DeStreak , 2% vol/vol isoelectric focusing buffer pH 4–7 , and trace amounts of bromophenol blue ) to a final volume of 250 μL . Immobilized pH gradient strips ( pH 4–7 , 13 cm; GE Healthcare ) were rehydrated at 30 V for 12 h . Isoelectric focusing was performed at 20°C using a Multiphor III system ( GE Healthcare ) as follows: 200 V for 2 h , 500 V for 2 h , 1000 V for 5 h , and a gradient applied from 1000 to 5000 V for 2 h . Finally , the voltage was set to 5000 V for 60 , 000 Vhr . After 1D isoelectric focusing , the IPG strips were reduced for 15 min with 1 . 5% dithiothreitol and alkylated for 15 min with 2 . 5% iodocetamide in equilibration buffer ( 6 M urea , 50 mM Tris-HCl pH 6 . 8 , 30% glycerol , and 2% sodium dodecyl sulfate ) . Second-dimension separation was carried out by placing equilibrated IPG strips onto 10 % polyacrylamide gels , sealing them with 0 . 5 % [wt/vol] low-melting-point agarose , and separating the proteins at 10°C using a Hoefer SE 600 unit ( 15 mA/gel for 30 min and then 23 mA/gel until the dye front reached the bottom of the gel ) . Proteins were developed with silver staining [51] or were directly transferred for immunoblotting . For 2D immunoblotting , proteins were transferred onto 0 . 45-μm polyvinylidenedifluoride membranes at 25 V for 1 h with transfer buffer [52] using the Trans-Blot SD semi-dry system . The success of electrotransference was evaluated by staining with 0 . 15% Ponceau S and 1% acetic acid 1% [vol/vol] . Membranes were destained and free binding sites were blocked overnight in phosphate-buffered saline blocking buffer ( 1% bovine serum albumin supplemented with 0 . 05% [vol/vol] Tween 20 , 5% [wt/vol] skim milk , pH 7 . 6 ) at 4°C . Membranes obtained from 2D gels were probed with 1:500 primary antibody ( gold standard pooled feline sera; n = 10 ) under the conditions used for 1D immunoblotting . Immunoreactive antigens were detected using an enhanced chemiluminescence detection kit ( GE Healthcare ) . 2D immunoblots were imaged using the method used for 1D immunoblots . Diagnostic values included sensitivity , specificity , positive predictive value , and negative predictive value . Receiver operating characteristic ( ROC ) curves were prepared and analyzed to determine the sensitivity and specificity of each antigen preparation for ELISA . The area under the curve ( AUC ) for ROC analysis was calculated to evaluate the diagnostic value of ELISA . We assumed that a test lacked diagnostic power when the ROC curve was linear with an AUC of 0 . 5 ( the ROC curve coincided with the diagonal ) . A powerful test was assumed to yield an AUC of ~1 . 0 , indicating the absence of both false-positives and false-negatives ( the ROC curve reached the upper left corner of the plot ) . To measure the degree of concordance of the results from preparations from strains CBS 132990 , CBS 132021 , CBS 132974 , and CBS 132984 , we calculated the kappa statistic and its 95% confidence interval ( CI ) . Kappa values were interpreted as follows: 0 . 00–0 . 20 , poor agreement; 0 . 21–0 . 40 , fair agreement; 0 . 41–0 . 60 , moderate agreement; 0 . 61–0 . 80 , good agreement; 0 . 81–1 . 00 , very good agreement [53] . P-values ≤0 . 05 were considered statistically significant . All calculations were performed with MedCalc Statistical Software version 14 . 8 . 1 ( MedCalc Software bvba; http://www . medcalc . org ) . Findings are reported in line with the STARD checklist for studies of diagnostic accuracy ( S1 Checklist ) .
We previously reported a high prevalence of S . brasiliensis in feline sporotrichosis outbreaks [8 , 9 , 20] . Based on this information , the main goal of the present investigation was to evaluate the presence and diversity of serum-derived antibodies against S . brasiliensis antigens in naturally infected cats . Further , we previously proposed the existence of a convergent humoral response in human sporotrichosis against antigens from S . brasiliensis , S . schenckii , and S . globosa [45] . To establish whether S . brasiliensis and S . schenckii express different antigens , we assessed whole cellular protein extracts from two strains of S . brasiliensis plus two strains of S . schenckii s . str . that were previously characterized by molecular [8 , 9 , 25 , 54] and serological [3 , 44–47] methods . Remarkably , and in support of our hypothesis that immunological distance increases with phylogenetic distance , sera from these cats reacted similarly , with no significant differences in titer between ELISA plates coated with proteins from S . brasiliensis or S . schenckii ( Fig 1 ) . ELISA detection of the four antigen preparations exhibited similar results , medians , and ranges for cats infected with sporotrichosis ( n = 49 ) : S . brasiliensis CBS 132990 , median 1 . 313 OD , 95% CI 1 . 262–1 . 489 OD; S . brasiliensis CBS 132021 , median 1 . 632 OD , 95% CI 1 . 462–1 . 714 OD; S . schenckii CBS 132974 , median 1 . 296 OD , 95% CI 1 . 157–1 . 442 OD; and S . schenckii CBS 132984 , median 1 . 028 OD , 95% CI 1 . 027–1 . 294 OD ( S1 Table ) . When using the assay to diagnosis cats with sporotrichosis , the area under the ROC curve was 1 . 0 ( 95% CI 0 . 94–1 . 000; P<0 . 0001; Fig 2 ) , indicating excellent performance . The control group of 19 non-Sporothrix infected animals was associated with lower medians and smaller ranges: S . brasiliensis CBS 132990 , median 0 . 2640 OD , 95% CI 0 . 2592–0 . 3098 OD; S . brasiliensis CBS 132021 , median 0 . 2590 OD , 95% CI 0 . 2517–0 . 2942 OD; S . schenckii CBS 132974 , median 0 . 2730 OD , 95% CI 0 . 2512–0 . 3136 OD; and S . schenckii CBS 132984 , median 0 . 2670 OD , 95% CI 0 . 2567–0 . 2907 OD ( S1 Table ) . Differences between the absorbance values for the infected and non-infected groups were statistically significant ( P<0 . 0001 ) . Sera from cats with other infections were non-reactive . Similar cutoff values yielded 100% specificity and sensitivity: S . brasiliensis CBS 132990 , 0 . 377 OD; S . brasiliensis CBS 132021 , 0 . 363 OD; S . schenckii CBS 132974 , 0 . 407 OD; and S . schenckii CBS 132984 , 0 . 346 OD ( S1 Fig ) . ELISA results showed very good agreement for the antigens assayed ( kappa = 1 . 0 ) . To diagnosis feline sporotrichosis via ELISA , we recommend the use of antigen preparations of S . brasiliensis , since this is the most prevalent species in feline sporotrichosis outbreaks . Antigen diversity was assayed with 1D immunoblots using the four antigen preparations of S . brasiliensis and S . schenckii tested via ELISA . Proteins extracts were evaluated according to the amount of protein extracted , the diversity of bands , the integrity of the samples , and the reproducibility of extraction . Approximately 2 μg of Sporothrix yeast whole-cell extracts were resolved by SDS-PAGE; silver staining revealed numerous proteins ranging from 10 kDa to 160 kDa in size , with different intensities . The Tris-Ca2+ extraction protocol [45 , 47] was suitable for the study of Sporothrix antigenic molecules during feline sporotrichosis , yielding samples with high amounts of protein and no degradation . As expected , antibodies from cats with sporotrichosis reacted with a wide variety of S . brasiliensis and S . schenckii proteins 20kDa to >160kDa in size ( Fig 3 ) . Cat-to-cat variation resulted in characteristic banding patterns for each animal ( Fig 3 ) ; supporting the hypothesis that in a genetically diverse population , the antibody repertoire is expected to vary among individual cats . On the other hand , we detected minor or no differences in IgG-reacting banding patterns between antigen preparations ( Fig 3 ) , consistent with the close genetic distance between S . brasiliensis and S . schenckii [8 , 9] . Despite this variation , all cats produced antibodies against a 60-kDa molecule in the S . brasiliensis proteome and a 70-kDa molecule in the S . schenckii proteome . The major antigenic S . brasiliensis molecules ( CBS 132990 and CBS 132021 ) recognized by feline IgG consisted of the following sizes: 60 kDa ( 100% and 100% , respectively ) , 90 kDa ( 92% and 92% , respectively ) , 100 kDa ( 86% and 86% , respectively ) , 38 kDa ( 60% and 56% , respectively ) , 40 kDa ( 56% and 58% , respectively ) , 45 kDa ( 44% and 42% , respectively ) , 30 kDa ( 36% and 26% , respectively ) , 52 kDa ( 30% and 32% , respectively ) , and 110 kDa ( 28% and 30% , respectively ) ( Fig 4A and 4C ) . Minor molecules recognized by feline IgG had sizes of 80 kDa , 25 kDa , 28 kDa , 120 kDa , 160 kDa , 35 kDa , 20 kDa , 55 kDa , 85 kDa , and 23 kDa ( Fig 4A and 4C ) . The major antigenic S . schenckii molecules ( CBS 132974 and CBS 132984 ) recognized by feline IgG had sizes of: 70 kDa ( 100% and 100% , respectively ) , 90 kDa ( 86% and 88% , respectively ) , 100 kDa ( 76% and 82% , respectively ) , 38 kDa ( 74% and 62% , respectively ) , 40 kDa ( 64% and 56% , respectively ) , 52 kDa ( 58% and 56% , respectively ) , 30 kDa ( 50% and 34% , respectively ) , 55 kDa ( 48% and 48% , respectively ) , and 45 kDa ( 40% and 30% , respectively ) ( Fig 4B and 4D ) . The minor S . schenckii molecules recognized by feline IgG had sizes of 25 kDa , 80 kDa , 28 kDa , 110 kDa , 120 kDa , 23 kDa , 35 kDa , 160 kDa , 85 kDa , and 20 kDa ( Fig 4B and 4D ) . Sera from uninfected cats did not react with S . brasiliensis or S . schenckii antigens . Sera from cats with other infections were also non-reactive in the immunoblot assay . The frequencies at which Sporothrix molecules were recognized in the antigen preparations are presented in S2 Table . There was no association between the number of bands recognized by each serum and the distribution and number of skin lesions on cats with sporotrichosis . We previously reported that the 60-kDa and 70-kDa proteins in S . brasiliensis and S . schenckii , respectively , are related to virulence profiles and are the main antigenic molecules during murine [3] and human [45] sporotrichosis . We also determined that this protein undergoes post-translational modification and is present as isoforms and glycoforms in the S . brasiliensis and S . schenckii proteomes [45] . We therefore investigated whether antibodies present in cat sera recognize all six isoforms in the S . brasiliensis proteome , as previously shown with human antibodies [45] . S . brasiliensis proteins were therefore resolved via 2D electrophoresis and immunoblotted with pooled sera from cats with sporotrichosis ( n = 10 ) and optimal antibody titers according to ELISA . Serum-derived antibodies in naturally infected cats mainly recognized all six isoforms of gp60 ( Fig 5 ) . The present results confirm that S . brasiliensis 3-carboxymuconate cyclase is a highly polymorphic protein [45] with sizes of 55–62 kDa and with isoelectric points of 4 . 45–4 . 80 . 2D immunoblotting revealed less diversity and more weakly reacting spots than 1D immunoblotting , perhaps due to protein loss during sample preparation for 2D electrophoresis ( compared to the crude extracts used in 1D immunoblotting and ELISA ) and serum dilution during pooling .
Gaining insight into host-parasite interplay in the immunological context is essential for understanding the emergence of feline sporotrichosis and is critical to serodiagnosis and the development of vaccines . Here , we demonstrated that antigens derived from yeast cell extracts of S . brasiliensis and S . schenckii s . str . yielded excellent results in ELISA and immunoblotting . The variety of molecules recognized by sera may be related to certain characteristics of the isolate , such as virulence , or even related to immune-system activation in each individual host . During infection , Sporothrix antigens elicit an IgG-mediated response; 3-carboxymuconate cyclase ( gp60 in S . brasiliensis and gp70 in S . schenckii ) is the immunodominant molecule in feline sporotrichosis , similar to murine and human disease . Therefore , this molecule may also be useful as a marker in the diagnosis of feline sporotrichosis and is a promising candidate for the development of therapeutic vaccines to tackle sporotrichosis in highly endemic areas . | Sporotrichosis is a neglected fungal disease of humans and animals that remains a serious public-health problem . Sporothrix infections persist in cats , leading to continued transmission via cat-to-cat and cat-to-human contact . Cats are the major source of transmission of Sporothrix brasiliensis to the human population . We stress the importance of implementing health policies aimed at detecting Sporothrix infection in cats as an attempt to reduce massive zoonotic transmission to humans . Early diagnosis of feline sporotrichosis is critical to recognize outbreaks areas and effectively tackle future spread of the disease among humans . We explored the diversity of molecules that are expressed by S . brasiliensis and S . schenckii and that are recognized by immunoglobulin G . Upon infection , the cat delivers an IgG-mediated response against Sporothrix antigens , similar to the response in murine and human sporotrichosis . We detected remarkable cross-reactivity among S . brasiliensis and S . schenckii antigens , supporting the hypothesis that antigenic epitopes may be conserved among closely related species . One protein , 3-carboxymuconate cyclase , was prominent in immune profiles from infected animals , using both types of Sporothrix antigens . Knowledge of the immune response in feline sporotrichosis is critical to advancing techniques for serological diagnosis , developing vaccines , and improving our understanding of Sporothrix evolution . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Proteomics-Based Characterization of the Humoral Immune Response in Sporotrichosis: Toward Discovery of Potential Diagnostic and Vaccine Antigens |
CDH11 gene copy number and expression are frequently lost in human retinoblastomas and in retinoblastomas arising in TAg-RB mice . To determine the effect of Cdh11 loss in tumorigenesis , we crossed Cdh11 null mice with TAg-RB mice . Loss of Cdh11 had no gross morphological effect on the developing retina of Cdh11 knockout mice , but led to larger retinal volumes in mice crossed with TAg-RB mice ( p = 0 . 01 ) . Mice null for Cdh11 presented with fewer TAg-positive cells at postnatal day 8 ( PND8 ) ( p = 0 . 01 ) and had fewer multifocal tumors at PND28 ( p = 0 . 016 ) , compared to mice with normal Cdh11 alleles . However , tumor growth was faster in Cdh11-null mice between PND8 and PND84 ( p = 0 . 003 ) . In tumors of Cdh11-null mice , cell death was decreased 5- to 10-fold ( p<0 . 03 for all markers ) , while proliferation in vivo remained unaffected ( p = 0 . 121 ) . Activated caspase-3 was significantly decreased and β-catenin expression increased in Cdh11 knockdown experiments in vitro . These data suggest that Cdh11 displays tumor suppressor properties in vivo and in vitro in murine retinoblastoma through promotion of cell death .
Retinoblastoma is initiated by loss of both RB1 alleles , denoted M1 and M2 mutational events [1] . These initiating events are sufficient for the development of the benign tumor , retinoma , but not enough to drive to malignancy; additional mutational events ( M3-Mn ) are required for development to retinoblastoma [1]–[3] . Early cytogenetic analysis performed on human retinoblastoma samples revealed recurrent chromosomal abnormalities [4] . Five comparative genomic hybridization ( CGH ) studies and one matrix CGH study confirmed these results and identified common genomic regions of gain and loss in retinoblastoma tumors [5]–[10] . Based on the location of these genomic changes , potential oncogenes ( KIF14 , E2F3 and DEK ) and tumor suppressor genes ( p75NTR and CDH11 ) have been identified and shown to have involvement in retinoblastoma development and progression [2] , [11]–[13] . Based on the frequency of and correlation between these mutational events , we proposed a genetic cascade to malignancy . Subsequent to loss of RB1 , the most frequent event is gain of 1q ( involving KIF14 ) [11] , followed by gain of 6p ( involving E2F3 and DEK ) [12] , [14] , and then , loss of 16q ( involving CDH11 ) [13] or gain of MYCN [15] . Understanding the pathway to tumorigenesis is important for the development of new and better therapeutics that can ultimately be used to halt retinoblastoma progression at an early stage . Importantly , delineating the order of mutational events in retinoblastoma , the prototypical model of cancer , is pertinent to the understanding of oncogenesis in general . In previous work , we narrowed the minimal region of genomic loss on chromosomal arm 16q22 . 1 to CDH11 [13] . This gene was lost in 58% of 71 retinoblastoma tumors , and its expression showed gradual loss in tumors of the murine retinoblastoma model ( TAg-RB ) induced by simian virus 40 large T-Antigen ( TAg ) expression [16] , with some advanced tumors ( 3 of 8 ) showing loss of Cdh11 . Thus , we proposed that Cdh11 acts as a tumor suppressor gene in retinoblastoma . Gratias et al , 2007 , identified a complex pattern of 16q loss of heterozygosity ( LOH ) in 18 out of 58 retinoblastoma samples . One tumor showed LOH at 16q24 , the region where CDH13 is located; however , CDH13 did not show reduced expression in retinoblastoma tumors , confirming our previous findings [13] , [17] . Gratias et al , 2007 did not test markers for CDH11 directly , as it was outside their minimal region of loss . They also correlated 16q allelic loss with diffuse intraocular seeding , implicating 16q loss as a late mutational event , in agreement with our proposed sequence of mutational events described in Bowles et al , 2007 [15] . Laurie et al . , 2009 , recently reported that loss of Cdh11 correlated with optic nerve invasion using an in vivo model of in vitro cell lines derived from an in vivo murine retinoblastoma model [18] . In our present study , we use the TAg-RB retinoblastoma mouse model to study the function of Cdh11 in tumorigenesis . This murine model , unlike any other RB mouse model , displays both molecular and histological features similar to the human disease [2] , [11]–[13] , [19] . Moreover , it is widely used as a pre-clinical model for testing therapeutics [14] , [20]–[25] . Perhaps the strongest resemblance to human tumors is evidenced by its initiation in the inner nuclear layer ( INL ) of the retina and the presence of Flexner-Wintersteiner rosettes . The latter is an important feature not recapitulated in any of the other mouse models of retinoblastoma [26]–[30] . We now address the roles of Cdh11 in developing retina and retinoblastoma . We report that Cdh11 is developmentally regulated during retinogenesis . We show that Cdh11 loss impacts the number of tumors that develop initially , and that it significantly increases the average tumor volume at PND84 per tumor initiating cell defined at PND8 in animals with mutant Cdh11 alleles with respect to animals with wild type alleles . We also show clear in vivo and in vitro evidence that more cell death occurs in tumors with wild type alleles than with mutant Cdh11 alleles , while cell proliferation remains unchanged regardless of Cdh11 allele status . Taken together , these data provide substantial evidence to suggest that in retinoblastoma Cdh11 acts as a tumor suppressor by facilitating cell death .
To assess the role of Cdh11 in the murine retina , we analyzed the spatio-temporal expression of cadherin-11 by immunostaining . Cadherin-11 was highly expressed by cells that typically differentiate at ED ( embryonic day ) 18 . 5 ( Figure 1A ) . At PND3 , expression was observed in areas where cells are migrating , and this became more visible in individual cells at PND6 . At PND15 ( data not shown ) and adult ( PND60 ) , cadherin-11 was expressed in the inner nuclear layer ( INL ) by Müller glia cell bodies and processes at the outer border of the outer nuclear ( ONL ) and inner border of the ganglion cell layer ( GCL ) ( Figure 1B ) . To identify retinal cell types that express cadherin-11 in the INL , we performed co-localization studies in adult retina , using retinal specific cell type markers . Cadherin-11 co-expressed with markers of horizontal cells and Müller glia cells and their processes , ( Figure 2A and 2B ) but not with markers of bipolar or amacrine cells ( Figure 2C and 2D ) . To examine the role of Cdh11 in the developing retina , we studied littermates of Cdh11 knockout animals . We analyzed retinas of Cdh11+/+ , Cdh11+/- , and Cdh11-/- on a 129/C57Bl-6 mixed background at developmental time points ED18 . 5 , PND3 , PND6 , PND15 and PND60 . To accurately compare the retina of varying genotypes , retinal sections were cut every 5 µm throughout the eyes in the papillary-optic nerve plane . Hematoxylin and eosin ( H&E ) analysis of retinal sections at all developmental time points revealed no gross phenotypic differences between the Cdh11 genotypes ( Figure 3 ) . Staining of retinal cell type markers was performed to determine if Cdh11 influenced differentiation . There was no obvious change in cell populations that expressed Chx-10 ( progenitor cells and bipolar cells ) , neurofilament ( 160 kDA for horizontal cells ) , cellular retinaldehyde-binding protein ( CRALBP for Müller glia cells ) or syntaxin ( HPC-1 for amacrine cells ) ( Figure S1 ) . The number of S-phase cells also seemed unaffected with loss of Cdh11 , determined by immunohistochemical analysis of BrdU positive cells ( Figure S1 ) . It is possible that the lack of gross phenotype in Cdh11-/- retinas is due to functional compensation by cadherins similar to Cdh11 . Cdh2 , also known as neuronal cadherin ( N-cadherin ) , shares 53% amino acid similarity to Cdh11 and is a mesenchymal cadherin like Cdh11 [31] . However , immunohistochemical analysis showed no change in expression of Cdh2 in the absence of Cdh11 ( Figure S1 ) . To evaluate cadherin-11 expression in developing tumors of the TAg-RB mouse model , we stained for cadherin-11 at PND9 , PND28 , PND35 , PND84 and PND140 . At PND9 , early initiating tumour cells showed complete overlap of TAg and cadherin-11 staining ( Figure 4A ) . At later time points , cadherin-11 expression was gradually lost from tumors: at PND28 , some tumors showed loss and others showed expression ( Figure 4B ) ; at PND35 , most tumors had lost expression of cadherin-11 ( Figure 4C ) ; by PND140 , large , late stage tumors showed complete absence of cadherin-11 expression ( Figure 4D ) . TAg-RB tumor development has been characterized ( unpublished data ) . At PND8 , TAg was first expressed by single cells in the INL of the retina ( Figure 5A ) . At PND28 , clusters of TAg-positive cells emerged ( Figure 6A ) , consistent with multifocal tumors , each derived from single TAg expressing cells already present at PND8 . These small tumor foci showed evidence of Homer Wright rosettes ( data not shown ) . At PND84 , tumors resembled human retinoblastoma ( Figure 7A ) . To examine the tumor suppressor role of Cdh11 in retinoblastoma development , we crossed Cdh11-/- mice with TAg-RB mice and analyzed genotypes Cdh11+/+;TAg+/- , Cdh11+/-;TAg+/- , and Cdh11-/-;TAg+/- , on a mixed 129/C57Bl-6 background . Gross phenotypes at varying time points were assessed by H&E staining . At PND8 , retinal histology of mice with normal and Cdh11 allelic losses showed no differences in H&E staining ( Figure 5A ) . Immunostaining showed that TAg was expressed by large , spindle shaped single cells in the INL ( Figure 5A ) . Tissue sections taken every 300 µm spanning the entire eye were manually counted for TAg-positive cells and the total number of TAg-positive cells per eye was extrapolated to the entire retina based on the total number of sections that were produced per eye ( Figure 5B , for detailed report of this method see [32] ) . A striking reduction in the number of TAg-positive cells was observed in retinas of mice with mutant Cdh11 alleles compared to mice with normal Cdh11 alleles . Animals of the Cdh11+/+;TAg+/- genotype had a mean of 6 , 417 TAg-positive cells per entire retina compared to 3 , 874 and 2 , 230 in Cdh11+/-;TAg+/- and Cdh11-/-;TAg+/- genotypes respectively , describing a significant allele dosage effect ( p = 0 . 01 , n = 5 ) ( Figure 5B ) . As a control and to normalize the total TAg-positive cells per retina , retinal area was measured for each of the selected sections , using the Image J software and then extrapolated to the entire retina . Total retinal areas at PND8 were found to be similar in all Cdh11 genotypes ( p = 0 . 83 , n = 5 ) ( Figure 5B ) . To quantify tumor-initiating cells with respect to retinal area , we determined the ratio of TAg-positive cells per retinal area , which showed a significant reduction correlated with Cdh11 genomic loss ( p = 0 . 01 ) ( Figure 5C ) . This effect continued at later stages in development , since at PND28 , fewer multifocal tumors developed in mice with Cdh11 loss . These data suggest that in this model , the expression of TAg may be dependent on Cdh11 . At PND28 , we observed a significant decrease in the number of multifocal tumors with decreasing number of functional copies of Cdh11 as assessed by both H&E and TAg stain ( Figure 6A ) . Tumor volumes as a percent of retina were estimated to be 5 . 0% , 3 . 2% and 1 . 5% in Cdh11+/+;TAg+/- , Cdh11+/-;TAg+/- and Cdh11-/-;TAg+/- genotypes respectively ( 5 animals analyzed per genotype ) . These analyses describe a significant decrease in tumor volume as Cdh11 alleles are lost ( p = 0 . 016 , Figure 6B ) . At PND84 , tumor morphology of the varying genotypes did not differ by H&E or TAg staining ( Figure 7A ) . All three genotypes showed tumors highly reminiscent of human retinoblastoma , presenting with large tumors originating from the INL , bulging into adjacent layers , and displaying features of Homer Wright rosettes ( Figure 7A ) . In stark contrast to earlier timepoints , total tumor volume at PND84 was not significantly different in mice of Cdh11+/+;TAg+/- , Cdh11+/-;TAg+/- and Cdh11-/-;TAg+/- genotypes ( p = 0 . 26; n = 8 , 8 , and 9 respectively , Figure 7B ) . However , unlike in the younger mice , total retinal size was significantly larger ( p = 0 . 01 ) in the Cdh11 null mice compared to Cdh11 normal mice ( Figure 7B ) , suggesting that loss of Cdh11 may affect the overall size of the adult retina in TAg mice . Tumor volume as a percentage of the entire retina was not significantly different between genotypes ( p = 0 . 07 , Figure 7C ) . The similarity of tumor volume at PND84 suggests faster tumor growth may be occurring in mice with mutant Cdh11 alleles , considering that fewer multifocal tumors were initially present at PND28 . These data suggest two roles for Cdh11 in retina: 1 ) Cdh11 displays tumor suppressor abilities in vivo and 2 ) Cdh11 loss affects retinal development in TAg mice , reflected in increase in overall size of the adult retina . This difference was not observed up to PND60 in Cdh11-/- mice ( Figure 3 ) . To establish whether tumors developing in Cdh11 mutant animals grew faster , we studied the rate of tumor growth between PND8 and PND84 . This was done by calculating the ratio of tumor volume at PND84 ( in pixels ) to the mean number of TAg-positive cells ( single tumor initiating cells ) at PND8 . The analysis revealed significant differences between the genotypes ( p = 0 . 003 , Figure 8A ) , indicative of faster growing tumors in mice with mutant Cdh11 alleles . We performed a second comparison to account for the difference in retinal size between genotypes at PND84 by calculating the ratio of percent tumor volume per retina at PND84 to the mean number of TAg-positive cells in the entire PND8 retina per genotype . Even after adjusting for retinal size , the tumor volume per initiating cell in mice with mutant Cdh11 alleles remained significantly greater ( p = 0 . 01 , data not shown ) . In addition , we noticed that while the retinal size at PND8 was similar between genotypes ( p = 0 . 83 , Figure 5B ) , the PND84 retinal size was significantly larger ( p = 0 . 01 , Figure 7B ) , suggesting a role for Cdh11 in the retinal development of TAg-RB mice . Since “growth” reflects a positive balance between cell proliferation and cell death we evaluated both cell proliferation and death in tumors at PND84 . At PND84 , tumors are well defined and easily quantifiable . We performed PCNA staining ( a marker of cells in early G1 and S phase ) of PND84 tumors in selected sections of Cdh11+/+;TAg+/- ( n = 2 ) and Cdh11-/-; TAg+/- ( n = 2 ) mice and calculated the percent PCNA positive cells per tumor volume revealing little difference between the two genotypes . To improve the power of this analysis , BrdU incorporation in PND84 tumors was evaluated in an additional larger cohort of animals . Again , we noticed no significant difference between the genotypes ( p = 0 . 121 , n = 6 for each genotype , Figure 8B ) . These data strongly support that Cdh11 is not acting to impede proliferation of tumor cells . To assess cell death , selected sections of Cdh11+/+;TAg+/- ( n = 8 ) and Cdh11-/-; TAg+/- ( n = 6 ) were manually counted for activated caspase-3 positive cells per tumor area and extrapolated to the entire tumor volume . Non-tumor retina showed no activated caspase-3 positive cells . We found significantly more cell death in tumors of mice with normal Cdh11 alleles than in tumors of mice with mutated Cdh11 alleles ( p = 0 . 04 , Figure 8C ) . Interestingly , β-catenin mRNA was upregulated in the Cdh11-/- TAg+/- mice relative to the Cdh11+/+Tag+/- mice ( Figure S2C ) . Furthermore , we observed a wide distribution of cell death among Cdh11+/+;TAg+/- mice ( mean = 2 . 90×10−03 , standard deviation ±2 . 08×10−03 ) compared to mice with mutant Cdh11 alleles ( mean = 6 . 94×10−04 , standard deviation ±7 . 25×10−04 , Figure 8C ) . To further support the role of Cdh11 in apoptosis , we assayed by immunohistochemistry , in an additional cohort of animals , five pro-apoptotic proteins: activated caspases 3 , 8 , 9 , TRAIL and BAX . Depending on the cell death marker , we observed 5 to 10 fold less expression in CDH11 mutant animals than in animals with normal Cdh11 alleles ( p<0 . 03 for all five cell death markers , Figure 8C ) . We also assessed cell death in vitro in a primary cell line derived from TAg-RB tumors ( T+539 ) . This tumor cell line , when treated with cadherin-11 siRNA , showed significant cadherin-11 knockdown ( Figure 8D ) . Following knockdown , caspase-3 expression was decreased ( Figure 8D ) , providing further evidence that Cdh11 acts to promote apoptosis . In addition , we studied RNA from the T+539 cell line treated either with Cdh11 siRNA or scrambled siRNA by RT-PCR for proliferation markers PCNA and Ki67 , and found no difference in expression ( Figure S3 ) . These data strongly support the hypothesis that Cdh11 has a pro-apoptotic role in TAg-RB tumors , but does not suggest a role in cell division or proliferation .
We have previously described copy number and expression loss of CDH11 in human retinoblastomas , suggesting a tumor suppressor role [13] . We now confirm the tumor suppressor role Cdh11 in retinoblastoma through functional experiments . The 97kD Cdh11 isoform that is retained in the Cdh11 knockout model we studied , has been documented to lack adhesion properties and thus likely represents a non-functional protein [33] , [34] . By crossing this CDH11 functional knockout with the TAg-RB mice , we report an unexpected result: Cdh11 allelic loss results in fewer tumor initiating TAg positive cells at PND8 ( Figure 5 ) , and consequently fewer multifocal tumors at PND28 ( Figure 6 ) compared to animals with normal Cdh11 alleles . This suggests that TAg transgene expression may be affected by the loss of Cdh11 ( Figure 5 ) . Loss of Cdh11 in Cdh11-/- mice did not affect retinal size up to PND60 ( Figure 3 ) . At PND8 retinal volumes were similar in Cdh11+/+;TAg+/- , Cdh11+/+;TAg+/- , and Cdh11+/+;TAg+/- mice , but at PND84 , the total retinal size was significantly larger ( p = 0 . 01 ) in the Cdh11-/-;TAg+/- mice compared to Cdh11+/+;TAg+/- mice ( Figure 7B ) , suggesting that loss of Cdh11 , when combined with the expression of TAg , affects the overall size of the adult retina in TAg-RB mice . Our previous studies of the Cdh11-/- retina , quantifying the individual retinal cell types visualized by immunofluorescence with cell-specific antibodies , showed no difference between the Cdh11-/- and wild type retina [35] . At PND84 , we show that absolute tumor volume was not statistically different between all three genotypes ( total tumor volume alone or as a percentage of the retinal volume ) . However , since these tumors arise from fewer tumor-initiating cells , we conclude that tumor growth per initiating cell was greater in mice with mutant Cdh11 alleles ( Figure 7B and 7C , Figure 8A ) . We conclude that Cdh11 functions as a tumor suppressor . Since tumor “growth” results from an imbalance between cell death and proliferation , we examined cell proliferation ( Figure 8B ) and cell death ( Figure 8C ) in TAg-RB tumors of mice with normal Cdh11 alleles versus mutated Cdh11 alleles . Our data indicate that when Cdh11 is lost , cell death is deficient while proliferation remains unchanged , suggesting that the tumor suppressor function of Cdh11 is mediated through promotion of apoptosis rather than inhibition of cell proliferation . This is further supported by our in vitro data showing significant decrease in caspase-3 and increase in β-catenin expression in Cdh11 knockdown experiments using siRNA ( Figure 8D and Figure S2A , S2B ) , while proliferation markers PCNA and Ki67 remain unchanged ( Figure S3 ) . The spread of tumor volumes across the various time points is narrowed in mice that have lost both Cdh11 alleles . We speculate that tumors in mice with normal Cdh11 alleles could be losing functional Cdh11 at varying timepoints during tumor development , and the wide spread in tumor volume reflects heterogeneity for Cdh11 . In contrast , mice with both Cdh11 alleles mutated have more consistent measures of cell death ( Figure 8C ) . This agrees with our previous report where some tumors display loss of Cdh11 , while others retain it at later timepoints [13] . In summary , we describe a mechanism by which Cdh11 may be functioning as a tumor suppressor gene in retinoblastoma . Additional experiments need to be performed to assess the mechanism by which Cdh11 facilitates cell death in these tumors . Our preliminary experiments have shown increased protein and mRNA levels of β-catenin when Cdh11 is knocked down , and increased β-catenin mRNA in PND84 Cdh11-/-TAg+/- mice relative to Cdh11+/+TAg+/- mice ( Figure S2A , S2B , S2C ) . Upon cell-cell contact , cadherin molecules form the adherens junction . The cadherin binds directly to β-catenin , which recruits α-catenin to link the complex to the cytoskeleton . This is necessary to maintain cell-cell adhesion and cellular architecture [36] . These junctions are dynamic and the structure and signaling provided by the complex ultimately determines the cellular phenotype and behavior [37] . β-catenin is additionally a major regulator of the Wnt signaling pathway . The Wnt-signaling pathway is implicated in other cancers [38] , [39] and suppresses apoptosis through both β-catenin dependent and independent pathways [40] . Many studies have shown that cadherin protein levels impact canonical Wnt-signaling and β-catenin levels . Gain and loss of function studies support cadherins directly sequestering β-catenin from the nucleus , acting as a sink for the cytosolic pool [41]–[43] . Additionally , downregulation of E-cadherin expression has been paralleled with an upregulation of β-catenin in hepatocellular carcinoma tumors [44] . Next investigations would test the possibility that down regulation of cadherin-11 affects the levels of canonical Wnt signaling in these TAg-RB cell lines , that may lead to the decrease in cell death and faster growing tumors . Previous studies of cell adhesion molecules in the neural retina have described that expression of cadherin subtypes is restricted to different retinal cell populations . Based on these studies the authors suggested that cadherins play a role in maintaining selective neuronal associations [45] , [46] . In order to understand the role of Cdh11 in retinoblastoma progression , we examined its presence during healthy retinal development . We showed that Cdh11 is developmentally regulated . Expression was restricted to differentiating/migrating retinal cells at E18 . 5 through to PND6 , and to the INL at PND60 ( adult ) ( Figure 1 ) . Cadherin-11 co-expresses with markers of Müller glia cell bodies and processes that span the entire retina ( Figure 1B and Figure 2B ) . Prominent expression of cadherin-11 by retinoblasts at PND3 and PND6 in the developing retina and co-expression with Müller glia and horizontal cell types , suggests roles for cadherin-11 in morphogenesis , such as cell migration , sorting or positioning of these cells ( Figure 1A ) during retinal development . The tumor -initiating cell in the TAg-RB mouse model has been identified to belong to a subset of the Müller glia ( unpublished data ) . Our results indicate that when Cdh11 alleles are mutated in TAg-RB mice , fewer cells express TAg and develop into retinoblastoma . It is possible that Cdh11 loss affects the expression of the TAg transgene in this murine model , or that it affects development of the subpopulation of Müller glia that gives rise to the TAg-RB tumours . We were unable to discern the latter , since Cdh11-/- mice do not show a significant change in retinal cell type distribution in the retina , and so few of this retinal subtype express TAg in this model ( unpublished data ) . From these data , we suggest that Cdh11 has an important role in the expression of TAg from the transgene in this murine model . We describe the use of the retinoblastoma TAg-RB mouse model to study specific gene function in tumor development . This was achieved by crossing TAg-RB mice to Cdh11-/- mice . We showed that Cdh11 is a suppressor of retinoblastoma progression by using a unique and highly sensitive method to identify and quantify tumor volume . Although fewer multifocal tumors initiate in mice with mutant Cdh11 alleles , suggesting that Cdh11 loss modulates the number of TAg-espressing cells in this murine model , the resulting tumors grow faster , describing a tumor suppressor role for Cdh11 in retinoblastoma progression . Significantly reduced numbers of cells stained for pro-apoptotic proteins in tumors of mice with absent Cdh11 alleles , indicating that promotion of cell death is an important part of the tumor suppressor action of Cdh11 .
All animals were maintained and sacrificed using protocols approved by the Animal Care Committee of the Ontario Cancer Institute ( OCI ) which adhere to the EC Directive 86/609/EEC for animal experiments . Cdh11-/- mice , background strain 129 , were provided by Dr . M . Takeichi [33] . To study the role of Cdh11 in retinal development , one-generation crosses were made between Cdh11-/- 129 and Cdh11+/+ C57-Bl-6 to get a mixed background of 129/C57Bl-6 . Littermates , Cdh11+/+ , Cdh11+/- and Cdh11-/- on this 129/C57Bl-6 , mixed background were sacrificed at developmental time points: embryonic day ( ED ) 18 . 5 , post-natal day ( PND ) 3 , PND6 , PND15 and PND60 . To analyze proliferating cells , pregnant mothers at ED18 . 5 , pups at PND3 and PND6 , and adults at PND84 were injected with bromodeoxyuridine ( BrdU ) reagent ( 5-bromo-2′-deoxyuridine and 5-fluoro-2′-deoxyuridine , 10∶1 , used at 1 ml reagent per 100 g body weight , Cat# 00-0103 , Lot# 60203722 , Zymed Laboratories ) for 2 hours and then sacrificed . TAg-RB ( TAg+/- ) , background strain C57/Bl-6 , mice were provided as a gift from Joan O'Brien [16] . One generation crosses were made between Cdh11-/- and TAg+/- mice to get double heterozygotes , Cdh11+/-; TAg+/- , on a 129/C56/Bl-6 mixed background . Mice were further crossed with Cdh11-/- , Cdh11+/- or Cdh11+/+ of a 129/C56Bl-6 mixed background to get the three genotypes analyzed for this study: Cdh11+/+;TAg+/- , Cdh11+/-;TAg+/- and Cdh11-/-;TAg+/- . These animals were sacrificed at PND8 , PND28 and PND84 , the latest time point we could study in compliance with our Animal Protocol at the Ontario Cancer Institute . Genotyping of Cdh11-/- mice and littermates were carried out using PCR conditions: 94°C , 2 min , 1 cycle , [94°C , 30 sec , 50°C , 30 sec , 72°C 30 sec] 30 cycles , 72°C 10 min , and 4°C cool block . Primers used were: forward , 5′ to 3′ ( 21 bp ) : ttc agt cgg cag aag cag gac and backward , 5′ to 3′ ( 19 bp ) : gtg tat tgg ttg cac cat g , and neo , 5′ to 3′ ( 23 bp ) : tct atc gcc ttc ttg acg agt tc . Sizes of expected PCR products were: Cdh11+/+: 240 bp , Cdh11+/-: 480 bp and 240 bp , and Cdh11-/-: 480 bp . Genotyping of TAg+/- mice and their littermates were carried out using similar PCR conditions: 94°C , 2 min , 1cycle , [94°C , 1 min 58°C , 1 min , 72°C 1 min] 30 cycles , 72°C 10 min , 1 cycle and 4°C cool block . Primers used were: forward 5′ to 3′: gac ttt gga ggc ttc tgg gat gca act gag and backward 5′ to 3′: ggc att cca cca ctg ctc cca ttc atc agt . Size of expected PCR product was 420 bp . Heads and/or eyes were fixed in freshly prepared 4% PFA/PBS for 48 hrs and then stored in 70% Ethanol . Heads were decalcified ( 8% formic acid following 4% PFA ) for approximately 1 week . Both heads and/or eyes were paraffin embedded and 5 µm sectioned . For Cdh11+/+ , Cdh11+/- and Cdh11-/- littermates: Serial sections were made specifically through the papillary-optic nerve plane ( approx . 20 sections in total ) for consistent comparison between genotypes . For Cdh11+/+;TAg+/- , Cdh11+/-;TAg+/- and Cdh11-/-;TAg+/- mice: Serial sections were made through the entire eye ( approximately 270–420 sections per eye with 5–7 sections made per slide ) . To estimate tumor volume per eye , we selected one slide every 60th section ( approx . one slide every 300 µm ) for analysis . A total of about 5–8 slides were analyzed per eye . Only one eye was analyzed per mouse . Slides selected for analysis were studied using the immunohistochemical protocol described previously[12] . Briefly , slides were incubated with primary , then biotinylated secondary antibodies , either anti-mouse , anti-rabbit , anti-goat , or anti-sheep , used at a dilution of 1∶200 with 10% DakoCytomation Antibody Diluent in 1% BSA/TBST for 1 hr at room temperature . To visualize TAg , BrdU and Brn3b ( ganglion ) stained cells , we employed an Immunopure DAB Substrate Kit ( Cat# 34065 , Pierce ) . After incubation with primary and biotinylated secondary antibodies , slides were incubated for 1 hr at room temperature in an ABC prepared solution ( Vectastain ABC Elite , Vector Laboratories ) . Stained cells could be visualized after a maximum of half an hour incubation in DAB substrate solution ( Pierce ) prepared fresh with 10% DAB/Metal Concentrate , 10× ( Product# 1856090 ) made in Stable Peroxide Substrate Buffer , 1× ( Product# 1855910 ) . All other proteins were visualized by immunofluorescence; after incubation with primary and secondary antibodies , slides were washed in 1×TBS and then incubated with Streptavidin-Alexa488 or Streptavidin-Alexa594 , used at 1∶200 , prepared in 1×TBS for 15 min at room temperature . Slides were washed briefly in 1×TBS and incubated in 4′ , 6-diamino-2-phenylindole ( DAPI ) used at 1∶50 , followed by wash in 1×TBS and mouniting with DakoCytomation Fluorescent Mounting Medium ( S3023 ) . Selected slides were Haematoxylin and eosin ( H&E ) stained for light microscopy analysis . Table 1 provides a complete list of all antibodies used . Antibodies to recognize specific cell types were: Hes-5 [47] , CRALBP [48] and glutamine synthetase [49] ( early Müller glia , Müller glia cell bodies and processes ) , syntaxin [50] ( HPC-1 for amacrine cells ) , neurofilament 160kDa [51] ( horizontal cells ) , Brn3b [52] ( ganglion cells ) and Chx-10[53] ( bipolar cells ) . Of the techniques described to measure tumor volume in murine retinoblastoma , none are useful to quantify small , developing tumors at very early time points [20]–[22] , [24] , [25] . Therefore , we developed a novel technique to quantitate tumor volume in the eyes of TAg-RB mice by analyzing every 60th section through the entire eye [32] . Tumor development was tracked by staining for TAg . Diamino benzidine ( DAB ) typically stains TAg cells brown with very little background , however in some cases , background staining is visible in the GCL and retinal pigment epithelium ( Figure 4 , Figure 5 , and Figure 6 ) . Total tumor area per eye was quantified as a percentage of total retinal area ( measured in pixels ) using Image J software . The selected sections were scanned at the Advanced Optical Microscopy Facility at the Ontario Cancer Institute using an Aperio ScanScope CS . Images were retrieved using ImageScope software and analyzed as a TIFF image using public domain image software: ImageJ: Image Processing and Analysis in Java available from http://rsb . info . nih . gov/ij/ . Retinas were manually traced for each eye and area was measured in pixels . For time point , PND8 , TAg positive cells in the retina were manually counted under a 40× inverted microscope ( Leica DMLB ) and for PND28 and PDN84 , the traced retinas were converted into an 8-bit format , and using a manually selected threshold tool , the tumor area ( DAB stained ) within the selected retina was highlighted and measured by the program in pixels . Total retina and tumor areas of all 5–8 analyzed sections per retina in one eye per animal were estimated calculating for percent tumor area [ ( tumor area in pixels/retina area in pixels ) * 100] . For PND8 , number of tumor cells per retinal area was used instead . BrdU positive cells were measured in pixels and quantified as an average/tumor area at PND84 . Positively stained apoptotic cells were also analyzed at PND84 and manually counted per section obtaining an average number per section . Five animals per genotype were analyzed at PND8 and PND28 . Seven animals of Cdh11+/+;TAg+/- genotype , eight animals of Cdh11+/-;TAg+/- genotype , and ten animals of Cdh11-/-;TAg+/- genotype , were analyzed at PND84 . The Kruskal-Wallis ( K-W ) Test was the main statistical method used to investigate differences in tumor and retinal size between the genotypes at various ages . Statistical analyses were performed using SAS version 9 . 1 ( SAS Institute , Cary , NC ) . All tests are two-sided and p-values equal or less than 0 . 05 were considered statistically significant . Cdh11 was knocked down in the TAg-RB derived cell line T+539 using three different stealth siRNAs: MSS202865 ( siRNA #1 ) , MSS202866 ( siRNA #2 ) and MSS202867 ( siRNA #3 ) ( Invitrogen Cat# 1320003 ) , using GL-2 vector siRNA ( Qiagen ) as a control . T+539 cells were transfected in triplicate with the siRNA at time of plating , using media without the addition of penicillin and streptomycin . The procedure included transfection of 125 pmol of each siRNA oligo in Lipofectamine 2000 ( Invitrogen ) , in a total of 2 ml plating medium . Cells were incubated for 24 hrs , 48 hrs , 72 hrs , 5 days , 7 days or 10 days . Knockdown was confirmed by immunoblot or RT-PCR for Cdh11 ( see below ) . Ideal inhibition was achieved 7 and 10 days post-transfection . RNA was isolated from the T+539 cell lines using the Trizol method . RNA was isolated from paraffin embedded tissue using modified GTC ( guanidine isothyocionate ) / proteinase K protocol . In short tissue was deparafinized through series of incubation in xylene and 100% ethanol followed by incubation in 1M GTC/6 mg/ml proteinase K solution for 6 hrs . GTC/proteinase K was removed by phenol extraction and RNA was precipitated by equal volume of isopropanol . Primers used for RT-PCR analysis were as follows: mCdh11: forward: 5′ atg agc ctc cca tgt tct tg 3′ , and reverse: 5′ggg tga tcg ctc tca cag at 3′; mKi67: forward: 5′ agc ctg tga ggc tga gac at 3′ , and reverse: 5′ ttt ctg cca gtg tgc tgt tc 3′; mPCNA: forward: 5′gaa ggc ttc gac aca tac cg 3′ , and reverse: 5′ cag cat ctc caa tgt ggc ta 3′; mTBP: forward: 5′ agc aac tgc agc agc ctc agt aca 3′ , and reverse: 5′ tct tcc tga atc cct tta aga tg 3′; mb-catenin: forward: 5′ caa gat gat ggt gtg cca ag 3′ , and reverse: 5′ ctg cac aaa caa tgg aat gg 3′ . Protein isolation and immunoblot analysis were performed as described previously [13] . Dilutions for cadherin-11 , caspase-3 and β-catenin antibodies used in immunoblot analysis are included in Table 1 . | Despite over two decades since loss of RB1 was implicated in initiating retinoblastoma , the unique tissue specificity of this process remains puzzling . Indeed , functional loss of both alleles of the RB1 tumor suppressor gene results in >40 , 000-fold increase in predisposition to retinal cancer during childhood , while one constitutional RB1 mutant allele confers a broader but much lower cancer predisposition later in life . We have proposed a specific signature of progressive genomic changes that leads to full tumor development . One of these changes is genomic loss of the CDH11 gene , suggesting that this gene normally suppresses the development of retinoblastoma . We present novel data indicating that Cdh11 functions as a tumor suppressor gene in retinoblastoma by facilitating cell death . Our insight into the sequence of events that contribute to retinoblastoma development is important for future therapies and fundamental understanding of cancer . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/neuronal",
"and",
"glial",
"cell",
"biology",
"oncology",
"ophthalmology/pediatric",
"ophthalmology",
"cell",
"biology/cell",
"signaling",
"cell",
"biology/cell",
"growth",
"and",
"division",
"developmental",
"biology/cell",
"differentiation",
"cell",
"biology/cell",
"adhesion",
"ophthalmology/ocular",
"tumors",
"cell",
"biology/gene",
"expression"
] | 2010 | Cdh11 Acts as a Tumor Suppressor in a Murine Retinoblastoma Model by Facilitating Tumor Cell Death |
CD8+ T-cell responses exert strong suppressive pressure on HIV replication and select for viral escape mutations . Some of these major histocompatibility complex class I ( MHC-I ) -associated mutations result in reduction of in vitro viral replicative capacity . While these mutations can revert after viral transmission to MHC-I-disparate hosts , recent studies have suggested that these MHC-I-associated mutations accumulate in populations and make viruses less pathogenic in vitro . Here , we directly show an increase in the in vivo virulence of an MHC-I-adapted virus serially-passaged through MHC-I-mismatched hosts in a macaque AIDS model despite a reduction in in vitro viral fitness . The first passage simian immunodeficiency virus ( 1pSIV ) obtained 1 year after SIVmac239 infection in a macaque possessing a protective MHC-I haplotype 90-120-Ia was transmitted into 90-120-Ia- macaques , whose plasma 1 year post-infection was transmitted into other 90-120-Ia- macaques to obtain the third passage SIV ( 3pSIV ) . Most of the 90-120-Ia-associated mutations selected in 1pSIV did not revert even in 3pSIV . 3pSIV showed lower in vitro viral fitness but induced persistent viremia in 90-120-Ia- macaques . Remarkably , 3pSIV infection in 90-120-Ia+ macaques resulted in significantly higher viral loads and reduced survival compared to wild-type SIVmac239 . These results indicate that MHC-I-adapted SIVs serially-transmitted through MHC-I-mismatched hosts can have higher virulence in MHC-I-matched hosts despite their lower in vitro viral fitness . This study suggests that multiply-passaged HIVs could result in loss of HIV-specific CD8+ T cell responses in human populations and the in vivo pathogenic potential of these escaped viruses may be enhanced .
Human immunodeficiency virus ( HIV ) induces persistent viremia leading to AIDS onset in humans . Virus-specific CD8+ T-cell responses exert strong suppressive pressure on HIV replication [1–3] but fail to control viremia in most infections . Several human leukocyte antigen ( HLA ) or major histocompatibility complex ( MHC ) alleles have been shown to be associated with lower viral loads [4–6] . Virus control associated with some of these protective MHC class I ( MHC-I ) alleles has been attributed to Gag epitope-specific CD8+ T-cell responses [6–9] . For instance , CD8+ T-cell responses specific for the HLA-B*57-restricted Gag240–249 TW10 and HLA-B*27-restricted Gag263–272 KK10 epitopes exert strong suppressive pressure on HIV replication , leading to lower viral loads [10–14] . Potent HIV-specific CD8+ T cells select for MHC-I-associated mutations resulting in viral escape from CD8+ T-cell recognition often with reduced in vitro viral fitness [15–18] . Virus transmission to MHC-I-mismatched individuals could result in reversion of these mutations to recover viral fitness [6 , 17 , 19–21] . Thus , it has been speculated that HIV may evolve by selection of individual MHC-I-associated mutations and their reversion after multiple transmissions among individuals with highly-diversified MHC-I genotypes . Recent studies have suggested that HIV evolves to have lower in vitro replication capacity through accumulation of MHC-I-associated mutations in human populations [17 , 22] . These studies in HIV-infected humans , however , have had difficulties in addressing the following issues . First , it is difficult to precisely trace serial HIV transmission . Second , it is difficult to compare in vitro viral fitness among highly-diversified HIV variants . Finally , it is difficult to evaluate the in vivo replication capacity of transmitted viruses . A macaque AIDS model of simian immunodeficiency virus ( SIV ) infection could be helpful to address these issues . We have previously established a group of Burmese rhesus macaques sharing individual MHC-I haplotypes [23 , 24] and reported the discovery of a protective MHC-I haplotype 90-120-Ia associated with lower setpoint viral loads after SIVmac239 infection [25] . SIV Gag206-216 and Gag241-249 epitope-specific CD8+ T-cell responses associated with 90-120-Ia are likely responsible for this reduction in viral loads [26 , 27] . SIVmac239-infected 90-120-Ia+ animals exhibiting persistent viremia consistently select Gag206-216 , Gag241-249 , Gag373-380 , Vif114-124 , Nef9-19 , Nef89-97 , and Nef193-203 epitope-specific CD8+ T-cell escape mutations by a year post-infection [28 , 29] . Two of these mutations , GagL216S leading to leucine-to-serine substitution at the 216th amino acid ( aa ) in Gag and GagD244E leading to aspartic acid-to-glutamic acid at the 244th , were shown to result in loss of viral fitness [26 , 27 , 30] . In the present study , we performed serial transmissions of SIV adapted to the protective MHC-I haplotype 90-120-Ia through MHC-I-mismatched rhesus macaques . To determine how viruses with 90-120-Ia-associated mutations can change after multiple transmissions , we first infected 90-120-Ia- macaques using a plasma sample from a 90-120-Ia+ macaque at 1 year post-infection with the SIVmac239 clone , and performed further plasma transmission through 90-120-Ia- macaques . Our analysis revealed that the viruses passaged through 90-120-Ia- macaques maintained 90-120-Ia-associated mutations and induced persistent viremia in 90-120-Ia- macaques despite their lower in vitro viral fitness . Notably , this passaged viral isolate showed rapid disease progression in 90-120-Ia+ macaques when compared to wild-type SIV . These results suggest that passaged viruses can maintain escape mutations and are thus less sensitive to CD8+ T cells restricted by protective MHC-I alleles .
We performed serial SIV transmissions in rhesus macaques . A plasma sample obtained from macaque #11 possessing the MHC-I haplotype 90-120-Ia 1 year after SIVmac239 infection was used as the first passage SIV ( 1pSIV ) ( Fig 1 ) . MHC-I Mamu-A and Mamu-B analysis detected only 90-120-Ia-derived alleles in this animal . Macaque #11 showed persistent viremia and developed AIDS 17 months after SIVmac239 infection [25] . The 1pSIV plasma sample was inoculated intravenously ( i . v . ) into 90-120-Ia- macaques #21 and #22 possessing MHC-I haplotypes 90-010-Ie and 89-002-Ip , respectively ( Fig 1 ) . Both animals showed persistent viremia ( Fig 2 ) and the second passage SIV plasma samples ( 2pSIV ) were obtained 1 year after 1pSIV infection from these two macaques ( 2pSIV1 from macaque #21 and 2pSIV2 from #22 ) . The 2pSIV1 and 2pSIV2 plasma samples were then inoculated i . v . into 90-120-Ia- macaques #31 and #32 possessing MHC-I haplotypes 89-002-Ip and 90-010-Ie , respectively ( Fig 1 ) . Both animals again showed persistent viremia ( Fig 2 ) and the third passage SIV plasma samples ( 3pSIV ) were obtained 1 year after 2pSIV infection from these two macaques ( 3pSIV1 from macaque #31 and 3pSIV2 from #32 ) . Thus , 3pSIV1 was obtained after transmissions through #11-#21-#31 while 3pSIV2 through #11-#22-#32 . No MHC-I alleles were shared among the former three animals but Mamu-B*066:01 was shared in two ( #22 and #32 ) of the latter three animals ( Fig 1 ) . Viral Gag- , Pol- , Vif- , Vpx- , Vpr- , Tat- , Rev- , and Nef-coding regions in 1pSIV had dominant mutations resulting in aa substitutions at twenty-nine residues ( 8 in Gag , 4 in Pol , 4 in Vif , 1 in Vpx , 1 in Vpr , 5 in Rev , and 6 in Nef ) as described previously [25] . These included seven mutations leading to aa substitutions at the 216th , 244th , and 375th residues in Gag , the 115th in Vif , and the 12th , 90th , and 201st in Nef , respectively . These replacements were previously shown to be selected for by CD8+ T cells in 90-120-Ia+ macaques by 1 year after SIVmac239 infection and result in escape from recognition by Gag206-216 , Gag241-249 , Gag373-380 , Vif114-124 , Nef9-19 , Nef89-97 , and Nef193-203 epitope-specific CD8+ T cells , respectively [28 , 29] . Of the twenty-nine mutations selected in 1pSIV , nine reverted in macaque #21 followed by one additional reversion and one re-selection in #31 . Six reverted in macaque #22 but two were selected again in #32 . Thus , twenty and twenty-five of twenty-nine mutations selected in 1pSIV remained in 3pSIV1 and 3pSIV2 , respectively ( Fig 3A and 3B and S1 Fig ) . Regarding the seven 90-120-Ia-associated CD8+ T-cell escape mutations described above , six remained without reversion even in 3pSIV1 and 3pSIV2 . The GagL216S mutation reverted in macaque #21 but was maintained in macaques #22 and #32 , while both 3pSIV1 and 3pSIV2 still had the GagD244E . Thus , the majority of 90-120-Ia-associated mutations remained without reversion even in 3pSIV through two passages . Macaques #21 , #22 , #31 , and #32 elicited CD8+ T-cell responses targeting multiple SIV antigens ( Fig 3C ) . All of these four animals exhibited high frequency Nef-specific CD8+ T-cell responses . Several mutations in addition to the twenty-nine selected in 1pSIV were selected in macaques #21-#31 and #22-#32 ( Fig 3A ) . Next generation sequencing ( NGS ) confirmed viral diversification in our transmitted plasma samples ( S2 and S3 Figs ) . Phylogenetic distances of viral Gag CA-coding region from wild-type SIVmac239 decreased in macaque #21 but increased in macaques #31 , #22 , and #32 , which may reflect the limited pressure exerted by Gag-specific CD8+ T-cell responses in #21 ( Fig 3C ) . Phylogenetic distances of Vif-coding and Nef-coding regions from wild-type SIVmac239 increased in individual animals . Changes in viral genome sequences were the largest in the Nef-coding region , possibly reflecting larger CD8+ T-cell responses targeting Nef ( Fig 3C ) . We then attempted to compare the in vitro replication capacity of wild-type SIVmac239 and the passaged viruses 1pSIV , 2pSIVs , and 3pSIVs . It is not easy to compare in vitro replication capacity of plasma HIVs directly and previous studies mostly used recombinant viruses derived from molecular clones such as NL4-3 where gag is replaced by the predominant plasma HIV sequences for comparison of in vitro viral fitness [17 , 22] . In the present study , we examined the in vitro replication capacity of viruses recovered from peripheral blood mononuclear cells ( PBMCs ) and plasma as well as recombinant SIVmac239-derived viruses whose gag was replaced by the predominant plasma SIV sequences . First , PBMCs from macaques #11 , #21 , #22 , #31 , and #32 at 1 year post-infection were cultured to obtain PBMC-derived virus stocks , referred to as c-1pSIV , c-2pSIV1 , c-2pSIV2 , c-3pSIV1 , and c-3pSIV2 , respectively . These viruses had the same nonsynonymous gag mutations with those in gag cDNAs amplified from plasma RNAs at 1 year post-infection ( Fig 4A ) . The culture supernatants of HSC-F cells ( a macaque T cell line ) on day 4 after infection with these PBMC-derived viruses showed lower reverse transcription ( RT ) activity compared to the wild-type SIVmac239 ( Fig 4B ) . These results indicate that all the PBMC-derived viruses , c-1pSIV , c-2pSIV1 , c-2pSIV2 , c-3pSIV1 , and c-3pSIV2 , have lower in vitro replicative capacities compared to wild-type SIV . Second , HSC-F cells infected with concentrated plasma samples obtained from macaques #21 , #22 , #31 , and #32 at 1 year post-infection were cultured to obtain the culture supernatants as passaged plasma-derived virus stocks , referred to as p-2pSIV1 , p-2pSIV2 , p-3pSIV1 , and p-3pSIV2 , respectively . We failed to recover a plasma-derived virus stock from macaque #11 . There were a few differences between p-3pSIV1-derived and plasma RNA-derived gag sequences , but p-2pSIV1 , p-2pSIV2 , and p-3pSIV2 had the same nonsynonymous gag mutations to those in gag cDNAs amplified from plasma RNAs at 1 year post-infection ( Fig 4A ) . Again , all of these virus-infected HSC-F cultures showed lower RT activity compared to the wild-type SIVmac239 in the culture supernatants on day 4 after infection ( Fig 4C ) . Furthermore , we compared in vitro viral fitness of these viruses with the wild-type SIV by competition assay . For comparison of wild-type and passaged viruses , HSC-F cells infected with individual virus stocks were cocultured to determine which viral genome sequences become dominant in the culture supernatants . In competition assay of wild-type SIVmac239 with any of the PBMC-derived virus stocks , the wild-type sequences became dominant ( S4 Fig ) . Competition assay using plasma-derived virus stocks confirmed the results obtained from the PBMC-derived virus stocks ( S4 Fig ) . These results indicate lower in vitro replication capacity of passaged viruses compared to the wild-type SIV . Finally , we constructed SIVmac239-derived recombinant viruses , SIV3p1gag and SIV3p2gag , where gag was replaced by the predominant 3pSIV1 and 3pSIV2 sequences , respectively ( Fig 4A ) . RT assay of the culture supernatants of HSC-F cells on day 4 after infection revealed lower in vitro viral fitness of both of these recombinant viruses compared to the wild-type SIVmac239 ( Fig 4D ) . Next , we used the 3pSIV1/2 plasma samples to challenge six 90-120-Ia+/90-010-Ie-/89-002-Ip- ( A[+]E[–]P[–] ) and six 90-120-Ia-/90-010-Ie-/89-002-Ip- ( A[–]E[–]P[–] ) macaques . Three A[+]E[–]P[–] macaques #411 , #412 , and #413 and three A[–]E[–]P[–] macaques #421 , #422 , and #423 were intravenously inoculated with 3pSIV1 , whereas three A[+]E[–]P[–] macaques #414 , #415 , and #416 and three A[–]E[–]P[–] macaques #424 , #425 , and #426 were intravenously inoculated with 3pSIV2 ( Fig 1 ) . All the 3pSIV-infected animals showed persistent viremia ( Fig 5A and 5B ) , despite the lower in vitro viral fitness of 3pSIV1/3pSIV2 . No clear difference was observed in viral loads between 3pSIV1 and 3pSIV2 infection in either A[+]E[–]P[–] or A[–]E[–]P[–] macaques . Furthermore , no significant difference was observed in viral loads between 3pSIV-infected 90-120-Ia- macaques and the SIVmac239-infected 90-120-Ia- control group ( n = 10 ) consisting of 90-010-Ie+ ( n = 6 ) and 90-120-Ib+ ( n = 4 ) animals that were previously reported to show typical levels of viremia [25] . Information on Mamu-A/B alleles in the control group was described in the previous report [25] . In contrast , 3pSIV-infected 90-120-Ia+ macaques exhibited significantly higher viral loads at week 3 post-infection compared with those in previously-reported wild-type SIVmac239-infected 90-120-Ia+ macaques ( n = 6 ) [25] ( p = 0 . 0087 by Mann-Whitney U-test ) ( Fig 5C ) . Interestingly , no significant difference in viral loads at week 1 was observed between these two groups ( Fig 5C and S5 Fig ) . Remarkably , 3pSIV-infected 90-120-Ia+ macaques showed significantly higher viral loads at months 4 . 5 , 6 , and 8 post-infection than the SIVmac239-infected 90-120-Ia+ macaques ( p = 0 . 0390 at month 4 . 5 , p = 0 . 0087 at month 6 , and p = 0 . 0087 at month 8 ) ( Fig 5C ) . This suggests that 3pSIV infection results in significantly higher setpoint viral loads in 90-120-Ia+ macaques than wild-type SIVmac239 does . 3pSIV infection showed significantly lower %CD4 at month 6 ( p = 0 . 0441 by Mann-Whitney U-test ) and shorter survival periods ( p = 0 . 0049 by Log-rank test ) than SIVmac239 in 90-120-Ia+ macaques ( Fig 6 ) . Indeed , three of six 3pSIV-infected 90-120-Ia+ macaques but none of the six SIVmac239-infected developed AIDS and had to be euthanized by a year post-infection , demonstrating that 3pSIV is more virulent than the wild-type SIVmac239 in 90-120-Ia+ macaques . One of the 90-120-Ia+ macaques , #412 , shared a 89-002-Ip-derived MHC-I allele Mamu-B*007:02 with macaque #31 . However , the five other 3pSIV-infected 90-120-Ia+ macaques ( excluding #412 ) still showed significantly higher setpoint viral loads and shorter survival periods than SIVmac239-infected 90-120-Ia+ macaques ( S6 Fig ) . In 90-120-Ia- macaques , no significant difference was observed in %CD4 nor survival periods between 3pSIV and SIVmac239 ( Fig 6 ) , although the former showed lower in vitro viral fitness when compared to the latter . The 90-120-Ia- macaques infected with 3pSIV with twenty or twenty-five of the twenty-nine mutations selected in 1pSIV showed one to five reversions and maintained no less than nineteen of them at 1 year post-infection ( Fig 7 ) . Regarding the seven 90-120-Ia-associated CD8+ T-cell escape mutations , no reversion was observed in three ( #422 , #423 , and #424 ) of the six animals , one in two ( #421 and #426 ) , and two in one ( #425 ) . On the other hand , the 90-120-Ia+ macaques infected with 3pSIV had no less than twenty-one of the twenty-nine mutations selected in 1pSIV at 1 year post-infection ( Fig 8A ) . Regarding the seven 90-120-Ia-associated CD8+ T-cell escape mutations , GagL216S was again selected for in all the three of the 3pSIV1-infected animals , and all seven mutations were dominant in the six 3pSIV-infected 90-120-Ia+ macaques except for #415 which showed a reversion at the 90th residue in Nef ( Fig 8A ) . Indeed , Gag206-216 , Gag241-249 , Gag373-380 , Vif114-124 , Nef9-19 , Nef89-97 , and Nef193-203 epitope-specific CD8+ T-cell responses were low frequency in these 90-120-Ia+ animals ( Fig 8B ) . Results suggest that these escape mutations in 3pSIV were maintained in 90-120-Ia+ macaques .
HIV induces persistent infection and accumulates viral mutations largely due to selection by CD8+ T cells . These mutations often have viral fitness costs and some of them can revert after viral transmission into MHC-I-mismatched hosts [6 , 17 , 19–21] . Recent studies in HIV-infected individuals have suggested that these MHC-I-associated mutations can accumulate in the population [22 , 31 , 32] . Analysis of HIV-infected transmission pairs has indicated that transmission of HIV mutations associated with the recipients' MHC-I alleles can result in higher viral loads [33 , 34] . Our present study in a macaque AIDS model demonstrated direct evidence indicating that MHC-I-adapted viruses that have been serially-passaged through MHC-I-mismatched hosts , even with lower in vitro viral fitness , can induce higher viral loads and more rapid disease progression in MHC-I-matched hosts . The majority of the mutations selected for in 90-120-Ia+ animal #11 were present in the 3pSIV virus isolate . Analysis of viral genome sequences in 3pSIV-infected 90-120-Ia- animals showed that the majority of the protective MHC-I haplotype 90-120-Ia-associated escape mutations were maintained after three serial passages . These mutations were preserved after three transmissions in 90-120-Ia- animals , supporting the notion that MHC-I-associated mutations can be maintained in circulating viruses in populations . The 90-120-Ia-associated mutations include GagL216S and GagD244E resulting in reduction of in vitro viral fitness [26 , 27] . Rapid reversion of these mutations was consistently observed after infection with SIV containing a single mutation in our previous study [35] . However , reversion occurred only rarely in infection with SIV carrying multiple mutations in the present study . No evidence of compensatory mutations which might have rescued viral fitness was found . After transmission , it may be more difficult for a virus with multiple prior CD8+ T-cell escape mutations to revert to wild-type when compared to a virus with a single escape mutation . The new host’s CD8+ T-cell response may exert the most important selective pressure on the multiply previously escaped virus ( selected for by the prior host’s CD8+ T cells ) and selection for the new host’s MHC-I-associated escape variants may occur first . These new CD8+ T-cell escape variants may have a greater selective advantage in vivo than any reversion of the previous host’s MHC-I-associated mutations and thus these changes will occur first perhaps delaying the reversion of the prior escape mutations . Previous studies examined the effect of viral genome mutations on in vitro viral fitness in the context of molecular HIV clones such as NL4-3 [17 , 22] . In the present study , we constructed recombinant viruses carrying 3pSIV-derived gag in the context of wild-type SIVmac239 and showed that these new recombinant viruses had lower in vitro replicative capacities when compared to the wild-type SIVmac239 . Mutations in gag appeared to have greater suppressive impact on in vitro viral fitness than those in other regions , consistent with previous reports [15–18] . Furthermore , we confirmed lower in vitro replication capacity of viruses recovered from plasma and PBMCs . Our results indicate that viruses carrying multiple MHC-I-associated mutations with lower in vitro viral fitness can be serially transmitted through MHC-I-mismatched hosts with maintaining the potential for higher viral loads in MHC-I-matched hosts . This suggests that MHC-I-adapted viruses can circulate in the population . 3pSIV obtained by serial passage through 90-120-Ia- macaques maintained several of the mutations selected for in the 90-120-Ia+ macaque #11 and induced higher viral loads and more rapid disease progression in 90-120-Ia+ hosts . These results demonstrate that MHC-I-adapted viruses can maintain the potential for higher virulence in MHC-I-matched hosts after serial transmissions through MHC-I-mismatched hosts . Although HIV transmissions from individuals with protective MHC-I alleles may be less efficient compared to those without protective MHC-I alleles , this study suggests that HIV isolates that are less sensitive to protective MHC-I alleles can be maintained and circulate in human populations . 3pSIV-infected 90-120-Ia+ macaques appeared to generate fewer mutations than wild-type SIVmac239 post-infection , and it is speculated that there were only a limited number of CD8+ T-cell targets in 3pSIV-infected 90-120-Ia+ macaques . This may be analogous to the situation in HIV-infected individuals where MHC-I homozygotes exhibit a more rapid course of disease progression [36] . Indeed , all of the 3pSIV-infected 90-120-Ia+ animals developed AIDS in 28 months post-infection , whereas 30–40% of 90-120-Ia- animals were alive without AIDS onset at 28 months after wild-type SIVmac239 or 3pSIV infection . In summary , we directly showed the impact of viral adaptation to MHC-I alleles on viral replication capacity in vivo . Protective MHC-I-adapted SIVs serially-passaged through MHC-I-mismatched hosts exhibited higher virulence in MHC-I-matched hosts despite their lower in vitro viral fitness . Our results indicate that MHC-I-adapted HIVs can circulate in populations , possibly resulting in loss of virus-sensitive MHC-I alleles in these populations .
Animal experiments were carried out in the Institute for Virus Research , Kyoto University ( IVRKU ) and Tsukuba Primate Research Center , National Institutes of Biomedical Innovation , Health and Nutrition ( NIBIOHN ) with the help of the Corporation for Production and Research of Laboratory Primates after approval by the Committee on the Ethics of Animal Experiments of IVRKU and NIBIOHN ( permission number: R13-11 , DS21-27 , DS23-19 , DS26-20 , and DS28-18 ) under the guidelines for animal experiments at IVRKU , NIBIOHN , and National Institute of Infectious Diseases in accordance with the Guidelines for Proper Conduct of Animal Experiments established by Science Council of Japan ( http://www . scj . go . jp/ja/info/kohyo/pdf/kohyo-20-k16-2e . pdf ) . The experiments were in accordance with the "Weatherall report for the use of non-human primates in research" recommendations ( https://royalsociety . org/topics-policy/publications/2006/weatherall-report/ ) . Animals were housed in adjoining individual primate cages allowing them to make sight and sound contact with one another for social interactions , where the temperature was kept at 25°C with light for 12 hours per day . Animals were fed with apples and commercial monkey diet ( Type CMK-2 , Clea Japan , Inc . ) . Blood collection and virus inoculation were performed under ketamine anesthesia . Animals were euthanized at the end of experiments or at the endpoint determined by typical signs of AIDS including reduction in peripheral CD4+ T-cell counts ( less than 200 cells/μl ) , 10% loss of body weight , diarrhea , and general weakness . At euthanasia , animals were deeply anesthetized with pentobarbital under ketamine anesthesia , and then , whole blood was collected from left ventricle . We performed serial SIV transmissions in Burmese rhesus macaque ( Macaca mulatta ) . A plasma obtained from macaque #11 possessing MHC-I haplotype 90-120-Ia at 1 year after SIVmac239 infection in our previous study [25] was used as the first passage SIV ( 1pSIV ) ( Fig 1 ) . In the present study , 0 . 2 ml of 1pSIV plasma was intravenously inoculated into 90-120-Ia- macaques #21 and #22 possessing MHC-I haplotypes 90-010-Ie and 89-002-Ip , respectively , and the second passage SIV plasma ( 2pSIV ) was obtained at 1 year after 1pSIV infection from these two macaques ( 2pSIV1 from macaque #21 and 2pSIV2 from #22 ) . Then , 0 . 2 ml of 2pSIV1 and 2pSIV2 plasma were intravenously inoculated into 90-120-Ia- macaques #31 and #32 possessing MHC-I haplotypes 89-002-Ip and 90-010-Ie , respectively , and the third passage SIV plasma ( 3pSIV ) was obtained at 1 year after 2pSIV infection from these two macaques ( 3pSIV1 from macaque #31 and 3pSIV2 from #32 ) . To investigate in vivo replication capacity of 3pSIV , 0 . 2 ml of 3pSIV1 was intravenously inoculated into three 90-120-Ia+/90-010-Ie-/89-002-Ip- ( A[+]E[–]P[–] ) macaques #411 , #412 , and #413 and three 90-120-Ia-/90-010-Ie-/89-002-Ip- ( A[–]E[–]P[–] ) macaques #421 , #422 , and #423 , while 0 . 2 ml of 3pSIV2 was intravenously inoculated into three A[+]E[–]P[–] macaques #414 , #415 , and #416 and three A[–]E[–]P[–] macaques #424 , #425 , and #426 ( Fig 1 ) . The data on six 90-120-Ia+ and ten 90-120-Ia- macaques intravenously infected with wild-type SIVmac239 ( Figs 5 and 6; S5 and S6 Figs ) were obtained in our previous study [25] . The determination of macaque MHC-I haplotypes was based on the family study in combination with the reference strand-mediated conformation analysis of Mamu-A and Mamu-B genes and detection of major Mamu-A and Mamu-B alleles by cloning the RT-PCR products as described before [24] . Confirmed MHC-I alleles consisting of MHC-I haplotypes 90-120-Ia , 90-010-Ie , and 89-002-Ip were described before [24 , 25] . Viral RNAs were extracted from plasma using the High Pure Viral RNA kit ( Roche ) . Fragments of cDNAs encoding SIVmac239 ( GenBank accession number M33262 ) Gag , Pol , Vif , Vpx , Vpr , Tat , Rev , and Nef were amplified from plasma RNAs by nested RT-PCR and subjected to direct sequencing by using dye terminator chemistry and an automated DNA sequencer ( Applied Biosystems ) as described before [28] . Predominant nonsynonymous mutations were determined . The Env-coding region known to have multiple antibody-related mutations was not included in the analysis . For pyrosequencing , cDNA fragments corresponding to nucleotides ( nt ) 1760–2463 ( containing entire Gag capsid [CA]-coding region ) , nt 5460–6340 ( containing entire Vif-coding region ) , and nt 9257–10167 ( containing entire Nef-coding region ) were used for making fragmentation libraries using GS FLX Titanium Rapid Library Preparation Kit ( Roche ) . The products were cleaned with Agencourt AMPure XP magnetic beads ( Beckman Coulter ) followed by quality control using Agilent 2100 Bioanalyzer ( Agilent Technologies ) . Emulsion PCR was performed with GS junior Titanium emPCR Kit Lib-L ( Roche ) . The emPCR products were deposited onto a GS Junior Titanium Pico Titer Plate and sequenced on the GS Junior System ( Roche ) . Sequencing reads were analyzed by the GS Amplicon Variant Analyzer Software ( Roche ) . After alignment of the FASTA files , populations of <1% were excluded . Molecular phylogenetic analyses were conducted by the Maximum Likelihood method using the MEGA6 software ( http://www . megasoftware . net/ ) . We recovered virus stocks from PBMCs and concentrated plasma samples obtained from macaques #11 , #21 , #22 , #31 , and #32 at 1 year post-infection . First , 1–5 x105 CD8- T cells negatively-selected from PBMCs were cultured in RPMI with 10% fetal bovine serum and 10 ng/ml human interleukin-2 ( hIL-2 ) ( Roche ) with stimulation by 2 μg/ml Phytohemagglutinin-L ( Sigma ) on the first 2 days . The culture supernatants on day 6 were added into HSC-F cells ( a cynomolgus macaque T-cell line ) [37] , which were cultured for 5–7 days to obtain the culture supernatants as PBMC-derived virus stocks . Second , plasma samples were concentrated by 6-fold using Lenti-X Concentrator ( Clontech ) and cocultured with HSC-F cells for 6–14 days to obtain the culture supernatants as plasma-derived virus stocks . To prepare the wild-type virus stock , we first obtained culture supernatants from MT4 cells ( a human T-cell line ) expressing CCR5 after transfection with the wild-type SIVmac239 molecular clone DNA ( pBRmac239 ) [38] . Then , these supernatants were added into HSC-F cells and the culture supernatants were obtained as the wild-type SIVmac239 stock , which was used for comparison of in vitro viral fitness with PBMC-derived and plasma-derived viruses . In addition , we constructed recombinant SIV clones by replacing the gag region ( nt 1056–3408 ) in the wild-type pBRmac239 molecular clone with that amplified from plasma RNAs of macaques #31 and #32 at 1 year post-infection . We then obtained recombinant molecular clones whose gag had the predominant 3pSIV1 and 3pSIV2 sequences , respectively . COS-1 cells were transfected with these molecular clones to obtain recombinant SIV3p1gag and SIV3p2gag virus stocks . COS-1 cells were transfected with pBRmac239 to obtain the wild-type SIVmac239 stock used for comparison of in vitro viral fitness with these recombinant viruses , SIV3p1gag and SIV3p2gag . Titers of these virus stocks were measured by RT assay as described previously [39 , 40] . For analysis of in vitro replication capacity , HSC-F cells were infected with these viruses ( 5 x 104 HSC-F cells were infected with viruses having the same RT activity with the wild-type SIVmac239 corresponding to 0 . 2 ng of p27 ) , and RT activity of the culture supernatants on day 4 post-infection was measured . In the competition assay for comparison of in vitro replication capacity of two kinds of viruses , HSC-F cells infected with individual virus stocks were cocultured to determine which viral genome sequences become dominant in the culture supernatants . HSC-F cells were infected with individual virus stocks ( normalized by RT activity ) and their coculture started next day . Coculture was continued by transferring the culture supernatant into fresh HSC-F cells every 4 days . RNA was extracted from the coculture supernatant and the Gag CA ( capsid ) -coding region was sequenced . When only one viral sequence became dominant on day 2 after the coculture initiation , we confirmed that the virus became dominant even in the coculture of the virus-infected cells with larger numbers of the other virus-infected cells in which both viruses were equivalently detected on day 2 . We measured antigen-specific CD8+ T-cell responses by flow cytometric analysis detecting gamma interferon ( IFN-γ ) induction after specific stimulation as described previously [41] . Autologous herpesvirus papio-immortalized B-lymphoblastoid cell lines ( B-LCLs ) were pulsed with individual SIVmac239 epitope-coding peptides ( at a final concentration of 1–5 μM ) or peptide pools ( at a final concentration of 1–2 μM for each peptide ) using panels of overlapping peptides spanning the entire SIVmac239 Gag , Pol , Vif , Vpx , Vpr , Tat , Rev , Env , and Nef amino acid sequences ( Sigma Aldrich Japan ) . PBMCs were cocultured with these pulsed B-LCLs under GolgiStop ( monensin , BD ) presence for 6 hours . Intracellular IFN-γ staining was performed with a CytofixCytoperm kit ( BD ) and fluorescein isothiocyanate ( FITC ) -conjugated anti-human CD4 ( M-T477 , BD ) , peridinin chlorophyll protein ( PerCP ) -conjugated anti-human CD8 ( SK1 , BD ) , allophycocyanin ( APC ) -conjugated anti-human CD3 ( SP34-2 , BD ) , and phycoerythrin ( PE ) -conjugated anti-human IFN-γ monoclonal antibodies ( 4S . B3 , Biolegend ) . Specific T-cell frequencies were calculated by subtracting non-specific IFN-γ+ T-cell frequencies from those after antigen-specific stimulation . Specific CD8+ T-cell frequencies lower than 0 . 02% of CD8+ T cells were considered negative . All statistical analyses were performed using Prism software ( GraphPad Software , Inc . ) with significance set at p values of < 0 . 05 . Comparisons were performed by Mann-Whitney U-test or log-rank test . | CD8+ T-cell responses exert considerable control over replication of HIV and select for viral escape mutations . Recent studies have suggested that these major histocompatibility complex class I ( MHC-I ) -associated mutations accumulate in populations and make viruses less pathogenic in vitro . Other studies have shown that some of these escape mutations can revert after passage to MHC-I-disparate hosts . In an attempt to reconcile these apparently conflicting results , we serially passaged a virus isolate through MHC-I-mismatched hosts in the macaque AIDS model of simian immunodeficiency virus ( SIV ) infection . Here we show an increase in the in vivo virulence of an MHC-I-adapted virus despite a reduction in in vitro viral replication capacity . Only a few of the selected escape mutations reverted after transmission to MHC-I-disparate recipients . Results clearly showed that MHC-I-adapted SIVs that have been serially-transmitted through MHC-I-mismatched hosts can have higher in vivo virulence in MHC-I-matched hosts despite their lower in vitro viral fitness . This study suggests that HIVs may become less sensitive to CD8+ T cell responses and could have increased in vivo virulence by adaptation to MHC-I in human populations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"microbial",
"mutation",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"immunology",
"microbiology",
"vertebrates",
"animals",
"mammals",
"genetic",
"mapping",
"retroviruses",
"primates",
"immunodeficiency",
"viruses",
"viruses",
"rna",
"viruses",
"cytotoxic",
"t",
"cells",
"viral",
"load",
"old",
"world",
"monkeys",
"white",
"blood",
"cells",
"monkeys",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"t",
"cells",
"siv",
"macaque",
"haplotypes",
"eukaryota",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"lentivirus",
"amniotes",
"organisms"
] | 2017 | In vivo virulence of MHC-adapted AIDS virus serially-passaged through MHC-mismatched hosts |
With the advent of subgenomic hepatitis C virus ( HCV ) replicons , studies of the intracellular steps of the viral replication cycle became possible . These RNAs are capable of self-amplification in cultured human hepatoma cells , but save for the genotype 2a isolate JFH-1 , efficient replication of these HCV RNAs requires replication enhancing mutations ( REMs ) , previously also called cell culture adaptive mutations . These mutations cluster primarily in the central region of non-structural protein 5A ( NS5A ) , but may also reside in the NS3 helicase domain or at a distinct position in NS4B . Most efficient replication has been achieved by combining REMs residing in NS3 with distinct REMs located in NS4B or NS5A . However , in spite of efficient replication of HCV genomes containing such mutations , they do not support production of infectious virus particles . By using the genotype 1b isolate Con1 , in this study we show that REMs interfere with HCV assembly . Strongest impairment of virus formation was found with REMs located in the NS3 helicase ( E1202G and T1280I ) as well as NS5A ( S2204R ) , whereas a highly adaptive REM in NS4B still allowed virus production although relative levels of core release were also reduced . We also show that cells transfected with the Con1 wild type genome or the genome containing the REM in NS4B release HCV particles that are infectious both in cell culture and in vivo . Our data provide an explanation for the in vitro and in vivo attenuation of cell culture adapted HCV genomes and may open new avenues for the development of fully competent culture systems covering the therapeutically most relevant HCV genotypes .
HCV is a positive strand RNA virus which belongs to the family Flaviviridae [1] . Its genome of about 9 . 6 kb is composed of the 5′non-translated region ( NTR ) , an open reading frame encoding a large polyprotein , and the 3′NTR [2] ( Fig . 1A ) . In the N-terminal region , the polyprotein is processed by cellular proteases to yield the structural proteins Core ( C ) , envelope proteins 1 and 2 ( E1 , E2 ) , and p7 . Cleavage of the non-structural ( NS ) proteins is accomplished by NS2 at the NS2/3 site and by the NS3 protease at all remaining sites [2] . NS4B induces cellular membrane alterations thought to provide a scaffold for the viral replication machinery [3] , [4] . NS5B is the viral RNA-dependent RNA polymerase whereas NS5A is an RNA binding phosphoprotein involved in RNA replication and virus assembly [5]–[8] . Two NS5A phospho variants have been described assumed to correspond to a basal and a hyper phosphorylated form ( p56 and p58 , respectively ) [9] . Phosphorylation of NS5A appears to be mediated by casein kinase I and II [7] , [10] . Interestingly , interference with NS5A hyperphosphorylation by inhibitors of casein kinase I such as H479 enhances viral RNA replication by more than 10-fold arguing that this modification is of disadvantage for high level replication [11] . About 170 million people are chronically infected with HCV . At present , neither a selective antiviral therapy nor a vaccine is available , and only a fraction of patients treated with a combination of polyethylene glycol ( PEG ) -conjugated interferon alpha ( IFN-α ) and ribavirin can be cured [12] . Thus , there is an urgent need for more effective antiviral treatment that also has fewer side effects . The development of such therapeutics and vaccines has long been hampered by the notoriously poor replication of HCV in cultured cells . The advent of subgenomic replicons that were originally derived from the genotype 1b isolate Con1 and that amplify efficiently in the human hepatoma cell line Huh-7 has in part overcome this limitation [13] . However , for Con1 mutations within the viral NS proteins are required to increase RNA replication to levels sufficient for experimental analyses [14]–[17] . These mutations have originally been designated ‘cell culture adaptive mutations’ , but should be renamed as ‘replication enhancing mutations’ ( REMs ) in order to discriminate them from cell culture adaptive mutations that increase virus titers without affecting replication [18]–[23] . The latter will therefore be designated ‘titer enhancing mutations’ ( TEMs ) throughout this report . REMs have been identified with all replicons derived from genotype 1 HCV isolates ( reviewed in [24] ) . These mutations cluster primarily in the center of NS5A and at distinct positions in NS3 and NS4B . Certain combinations of mutations enhance RNA replication cooperatively , the exact mode of action , however , remains elusive . The superior RNA replication capacity accomplished by adapted NS-proteins allowed the generation of efficiently replicating full length HCV genomes [25] , [26] . However , in spite of high level replication , these genomes do not or only very poorly support the production of detectable levels of infectious HCV particles [25] , [26] . Earlier we demonstrated that the presence of certain REMs interferes with infectivity in vivo [27] . A Con1 genome with a single mutation in NS5A ( S2197P ) was severely attenuated in chimpanzees inoculated intrahepatically with in vitro transcripts of this genome . A triple mutant carrying two mutations in NS3 ( E1202G , T1280I ) and the aforementioned substitution in NS5A was unable to establish a productive infection implying that REMs interfere with infectivity in vivo . In line with this assumption , efficient production of infectious HCV in cell culture so far has only been achieved with a genotype 2a HCV isolate ( designated JFH-1 ) that replicates to very high levels without requiring REMs [28] , [29] . Employing transient replication and virus release assays in this study we demonstrate that REMs interfere with the production of infectious HCV particles . We show that at least in case of the Con1 isolate , mutations especially in the NS3 helicase , but also in NS5A and NS5B lead to strong impairment of virus production whereas Huh-7 cells transfected with the wild type genome or a genome containing one REM in NS4B release substantial amounts of HCV that is infectious in cell culture and in vivo .
Given the infectivity of the wild type Con1 HCV isolate in vivo on one hand and its poor replicative capacity in transfected Huh-7 cells in vitro on the other hand , we first investigated whether this genome is capable of producing virus particles in cell culture . Four constructs were used for this experiment: Con1/wild type ( wt ) , a replication incompetent variant thereof with a mutation that destroys the active site of the polymerase ( Con1/D318N ) , a weakly adapted Con1 genome containing a REM in NS5A ( Con1/S2197P ) and a highly adapted Con1 genome containing two REMs in NS3 ( E1202T , T1280I ) and one in NS5A ( S2197P ) and designated Con1/NS3+S2197P ( Fig . 1A ) . We have shown earlier that the combination of these 3 REMs enhances RNA replication in a cooperative manner [15] In vitro transcripts were transfected into Huh-7 cells and RNA replication was quantified by Northern blotting and phospho imaging ( Fig . 1B and C , respectively ) , whereas virus production was measured by determining the release of core protein into the supernatant of transfected cells ( Fig . 1D ) . As expected Con1/wt replicated poorly in Huh-7 cells , displaying no clear difference to the replication incompetent mutant Con1/D318N . In contrast , elevated RNA levels were observed for the weakly and highly adapted genomes ( Con1/S2197P and Con1/NS3+S2197P , respectively ) . Analysis of extracellular core protein levels revealed that highest amounts of core were transiently released into the supernatant of cells transfected with Con1/wt ( Fig . 1D ) . Core protein was first detectable at 12 h post transfection , increased to peak levels at 24 h , and declined rapidly thereafter . This kinetic correlated well with intracellular RNA levels with highest amounts of Con1/wt input RNA detected right after transfection followed by a rapid decline due to the poor replication of this genome ( Fig . 1D ) . Interestingly , the same kinetic of core release was found with the replication inactive Con1/D318N mutant demonstrating that at least in this setting core release does not require RNA replication . Most importantly , the Con1 variants containing the REMs displayed clearly impaired core release . This defect was most prominent for the triple mutant Con1/NS3+S2197P which despite highest intracellular RNA and core quantities did not release core protein to detectable levels . Coherently , Con1/S2197P harbouring a single REM in NS5A exhibited impaired release of core relative to Con1/wt , although due to the adapted phenotype , intracellular RNA levels were elevated . In contrast , about 15% of intracellular core protein was released from cells transfected with Con1/wt ( Fig . 1E ) . The inverse correlation between increase of RNA replication by the REMs in NS3 and NS5A and the impaired core release suggested that such mutations interfere with virion production . To more quantitatively determine the correlation between the extent of replication enhancement by a given REM and the impairment of virus production we generated a panel of Con1-derived subgenomic reporter replicons and full length genomes into which single REMs were inserted . For this purpose we selected several representative REMs residing in NS3 ( E1202G , T1280I ) , or NS4B ( K1846T ) , or NS5A ( S2197P , S2204R ) , or NS5B ( R2884G ) that increase replication of subgenomic HCV replicons to various extents ( Fig . 2A ) [17] . In addition , we analyzed two different combinations of mutations in which the two REMs in NS3 were combined either with the REM in NS4B ( construct Con1/NS3+K1846T ) or NS5A ( construct NS3+S2197P ) . As shown in Fig . 2A amongst the single mutations , the REM located in NS4B ( K1846T ) enhanced RNA replication most whereas the two mutations residing in NS3 had least effects [15] , [17] . As described earlier , when these two REMs in NS3 were combined either with the mutation in NS4B or the S2197P substitution in NS5A , RNA replication was enhanced cooperatively ( replicons NS3+K1846T and NS3+S2197P , respectively ) [17] . To determine the impact of these mutations on core release they were inserted into the parental Con1 full length genome . Subsequently , mutants were transfected into Huh-7 cells and core release was determined by ELISA . The results in Fig . 2B show that strongest inhibition of virus production ( assembly or release ) was exerted by either of the two REMs in NS3 that enhanced RNA replication only to a minor extent . Profound impairment of core release was mediated by the REMs residing in NS5A or NS5B whereas the REM in NS4B still allowed core release to a level comparable with the wild type . However , when this mutation or the S2197P REM in NS5A was combined with the two mutations in NS3 core release was potently blocked . Given the rather efficient core release obtained with the Con1/K1846T genome , we performed a more detailed quantitative analysis by measuring the accumulation of intra- and extracellular core protein amounts at various time points post transfection ( Fig . 2C ) . In agreement with the elevated replication level , the amounts of intracellular core protein were consistently higher in cells transfected with Con1/K1846T as compared to Con1/wt transfected cells , but substantially lower compared to Con1/NS3+K1846T ( Fig . 2C , left panel ) . Likewise , amounts of core protein released into the supernatant of transfected cells were elevated in case of Con1/K1846T transfected cells ( middle panel ) . However , when correlating the amount of released core protein to the total amount of core protein expressed we found that also the NS4B mutation reduced core protein release , especially at later time points post transfection ( 48 and 72 h; Fig . 2C , right panel ) . In summary , our data suggest that REMs , at least those examined here , have a negative impact on core release , but the extent of this interference does not correlate with the extent of RNA replication enhancement . To confirm that core protein released into supernatants of cells transfected with Con1-derived constructs represents virus particles , we studied the requirements for core release by using reverse genetics . Several variants of the Con1 genome were generated: Mutant ΔE1-E2 comprising a large in frame deletion that encompasses most of the E1 and E2 coding sequence; mutant wt/A358Ins and K1846T/A358Ins in which an alanine codon was inserted into the E1 coding region at amino acid residue 358; mutants wt/NK367AA and K1846T/NK367AA in which two amino acid residues within the transmembrane domain of E1 were replaced by alanine residues . The A358Ins and NK367AA mutations were previously shown to disturb E1/E2 heterodimerization thereby blocking formation of functional glycoprotein complexes while the deletion of E1-E2 is known to abrogate release of infectious JFH1 particles [28] , [30] , [31] . Twenty four hours after transfection , total amount of core protein present in cell lysate and medium was determined by core-specific ELISA . Con1 wild type genomes expressed comparable amounts of core in the cell lysate arguing for comparable transfection efficiency ( Fig . 3A ) . Owing to higher replication , core amounts in cells transfected with the K1846T-constructs were higher . In case of the wild type up to about 45% of intracellular core protein was released , whereas in case of the K1846 mutant this value was reduced to about 20% ( Fig . 3C ) . Most importantly , core release was reduced to background levels ( as determined with the ΔE1/E2 mutant ) whenever the mutations in the envelope coding sequence were introduced and this effect was found both with the wt and the K1846T genome ( Fig . 3C ) . This result indicates that core release observed with these two genomes is a specific process that requires functional envelope glycoproteins . If authentic virus particles were produced from Con1/wt or Con1/K1846T transfected cells , the core protein shell which harbours the viral RNA should be surrounded by a lipid membrane containing the viral envelope glycoproteins . Consequently , core protein and HCV RNA should be captured by antibodies directed against E1 or E2 . To identify HCV-envelope specific antibodies to be used for Con1 virus capture assay we first screened a panel of human monoclonal antibodies recognizing HCV E2 or E1 proteins for their capacity to bind to HCV pseudo particles ( HCVpp ) generated in 293T cells by using an HIV-based vector and a HCV Con1 E1–E2 expression construct . The isotype-matched RO4 antibody which is directed against p64 of Cytomegalovirus ( CMV ) served as negative control . Upon incubation with monoclonal antibodies the amount of captured HCVpp was determined by using HIV p24-specific ELISA . As shown in Fig . 4A , highest yields were obtained with the human monoclonal antibodies CBH-5 , CBH-8C and CBH-2 [32] . For subsequent capture assays we used CBH-5 and CBH-2 , two efficient capture antibodies as well as CBH-7 that captured only low amounts of Con1-derived HCVpp and thus served as sensitivity control . The RO4 antibody was used as negative control . Concentrated cell culture supernatants derived from Con1/wt , Con1/K1846T and Con1/ΔE1/E2 transfected Huh-7 cells were incubated with antibody coupled beads and the quantity of captured HCV RNA was determined by qRT-PCR after extensive washing with PBS ( Fig . 4B ) . Although the overall capture efficiency was only about 5% , particles present in supernatant of Con1/wt and Con1/K1846T transfected cells could specifically be captured with all 3 CBH-antibodies . Capture efficiency correlated well with the one obtained with HCVpp . Only background amounts of RNA were captured from the supernatant with the non-specific antibody ( RO4 ) or from supernatant of Con1/ΔE1-E2 transfected cells demonstrating specificity of this capture assay . Biophysical properties of captured particles were further characterized by treatment with S7 micrococcal nuclease under various conditions . Assuming that intact particles protect the viral genome , we treated captured complexes with nuclease without or with prior treatment with the detergent Triton X-100 ( Fig . 4C ) . We found that the viral genome in captured complexes was resistant to nuclease treatment consistent with the protection of the RNA by an intact virus particle . Removal of lipids however , by detergent treatment rendered the viral genome fully nuclease sensitive and RNA levels after Triton X-100 and S7 treatment were at the background as determined with the capture with the R04 control antibody . Addition , of protease to the detergent-treated particles did not increase nuclease sensitivity ( not shown ) . These results suggested that HCV nucleocapsids –if they are formed at all- might be unstable after removal of the envelope and/or the lipoprotein shell . Densities of Con1-derived particles were determined by using density gradient centrifugation in comparison to virus particles produced from JFH-1 transfected Huh-7 cells and virus particles contained in a high titer patient serum . Gradient fractions were harvested from the top and core protein amounts contained in each fraction were determined by ELISA . We did not use qRT-PCR because of residual amounts of in vitro transcripts and plasmid DNA in virus preparations generated by RNA transfection precluding unambiguous measurements of particle associated HCV RNA . In the first set of experiments we compared particles derived from Con1/wt , JFH-1 and patient serum ( Fig . 5A ) . Most particles present in patient serum had a density of about 1 . 04 g/ml probably representing viruses associated with very low density and low density lipoproteins [33]–[36] . The minor peak with a density of ca . 1 . 12 g/ml may correspond to virus that is less complexed with lipoproteins . Interestingly , cell culture-derived Con1/wt and JFH-1 particles exhibited an analogous density profile but with very different ratios when compared to the patient-derived particles . Most particles had a density of 1 . 14 g/ml , whereas only a minor and broad peak was found at very low density in the range of 1 . 03 to 1 . 08 g/ml . In this respect , density profiles of Con1/wt and JFH-1 derived virus particles were indistinguishable . Likewise , density profiles of Con1/wt and Con1/K1846T derived particles were also very similar ( Fig . 5B ) arguing that the genome with this REM in NS4B is capable of producing virus particles , too . In summary , the similarity of buoyant densities of cell culture-derived Con1/wt , Con1/K1846T and patient-derived particles supported the notion that authentic HC virions were released from Huh-7 cells transfected with these HCV genomes . The different relative amounts of particles at very low and high densities may reflect differences of lipoprotein ‘imprinting’ of the particles by the host cell ( Huh-7 cells versus fully differentiated human hepatocytes ) in agreement with the reported defect of Huh-7 cells to produce vLDL [37] . Attempts to directly demonstrate infectivity of Con1/wt or Con1/K1846T derived particles in cell culture were complicated by the low replicative capacity of both genomes . However , recently Neddermann and colleagues demonstrated that inhibition of casein kinase I that appears to be responsible for hyperphosphorylation of NS5A , with compound H479 results in substantial enhancement of RNA replication of a non-adapted genome whereas replication of a genome containing a REM in NS5A was blocked [10] , [11] , [38] . Assuming that enhancing replication of Con1/wt and possibly also Con1/K1846T with H479 would facilitate detection of viral proteins in infected cells , we first established the optimal concentration of this kinase inhibitor required to stimulate replication of Con1/wt and Con1/K1846T in Huh7 . 5 cells . In agreement with the report by Neddermann and colleagues [11] we observed an about 5-fold increase of replication of Con1/wt at 48 h post transfection when Huh7 . 5 cells were treated with 10 µM of H479 whereas replication of Con1/K1846T was not enhanced ( Fig . 6A ) . Nevertheless , even under optimal conditions replication was about 10–100-fold below the level achieved with the highly cell culture adapted replicon Con1/NS3+K1846T ( data not shown ) . Having established the optimal conditions for H479-mediated enhancement of Con1/wt replication , we exploited this protocol to determine infectivity of HCV particles in tissue culture ( Fig . 6B ) . Huh7 . 5 cells , which are more permissive for HCV infection than Huh7-Lunet cells [39] were inoculated with about 200 Con1/wt or Con1/K1846T particles per cell ( calculated according to the amount of core protein and assuming 200 core protein molecules per particle ) or the analogous volume of concentrated supernatant of mock-transfected cells that was prepared in parallel . Cells inoculated with Con1/wt virus or mock-supernatant were treated with H479 whereas Con1/K1846T inoculated cells were left untreated . Three days after inoculation , cells were fixed and the replicase component NS5A was detected by immunofluorescence . As shown in Fig . 6C , under all conditions no NS5A-specific signal could be detected in cells inoculated with the control-supernatant . In contrast , a low but specific signal was found in Con1/wt inoculated cells and this signal was enhanced in cells that had been treated with the kinase inhibitor H479 . Likewise , NS3 and NS5A expression was detected in a few cells inoculated with Con1/K1846T containing supernatant ( Fig . 6D ) . Infection appeared to be a specific process because no signal was detected in Con1/wt inoculated cells that had been treated with Concanamycin A , which is an inhibitor of endosomal acidification and that was shown to block infection of Huh-7 cells with JFH-1 derived virus [40] ( Fig . 6C ) . Infection was also not detected upon infection of Huh7-Lunet cells that express low amounts of CD81 ( not shown ) . These results suggested that cell culture produced Con1/wt or Con1/K1846T particles are infectious in vitro . However , quantitative analyses could not be performed due to the low replication of these genomes . To firmly demonstrate infectivity of cell culture-produced Con1/wt and Con1/K1846T particles , we performed in vivo infection experiments . UPA+/+-SCID mice that had been xenografted with primary human hepatocytes , were inoculated with equally concentrated culture supernatants from Huh7 cells that had been transfected with Con1/wt or Con1/K1846T . As control we used supernatants from cells transfected with Con1/NS3+K1846T , which supports core release to only very low level . Final concentrations of HCV RNA in the purified and concentrated stocks were about 2×109 RNA copies ( IU ) per ml for all 3 preparations . Based on core-ELISA measurements this corresponded to 5 . 4×103 pg per ml of Con1/wt , 2×103 pg/ml of Con1/K1846T and 2 . 3×102 pg per ml core protein of Con1/NS3+K1846T , respectively . Assuming that one HCV particle contains about 200 copies of core protein ( an estimate that is derived from hepatitis B virus , which has a similar particle size [41] ) , the ( theoretical ) infection dose per animal ( 100 µl inoculum ) was about 7 . 7×107 particles in case of Con1/wt , 2 . 9×107 particles in case of Con1/K1846T and about 3×106 particles in case of Con1/NS3+K1846T , respectively . The comparable amounts of RNA detected in all 3 preparations argues that in case of the triple mutant RNA-containing replication complexes that contain no or very low amounts of core protein were released [25] . For each construct , two mice were inoculated with 100 µl of the concentrated stock and virus titers in serum were determined by qRT-PCR ( Fig . 7 ) . Unfortunately , one mouse inoculated with Con1/K1846T died spontaneously already at week 2 while the second mouse died at week 6 , presumably a follow-up reaction of serum withdrawal . The time course of infection shown in Fig . 7 demonstrates that mice inoculated with Con1/wt or Con1/K1846T particles were readily infected and remained viremic throughout the observation period . In case of the wild type , peak viremia was observed at week 3 post inoculation and steadily declined thereafter , most likely due to a decreased survival of the engrafted human hepatocytes . In contrast , mice inoculated with supernatants from Con1/NS3+K1846T transfected cells remained negative and viral RNA was not detected in any of the serum samples . The fact that already one week post inoculation these animals were RNA negative also shows that input virus did not interfere with the read-out . The mouse inoculated with Con1/K1846T virus also was readily infected and viremia was well detectable in all available serum samples . To confirm that the viral genome in the mouse inoculated with the Con1/K1846T mutant had not reverted to wild type , the serum sample obtained at week 6 post inoculation was used for cloning of a genome fragment covering the NS4B coding region . Sequence analysis confirmed that the HCV genome in this mouse had retained this particular mutation ( data not shown ) . In summary , these results convincingly demonstrate that Con1/wt and Con1/K1846T transfected Huh-7 cells release infectious HCV particles . Their production is blocked by various REMs , most notably those residing in the NS3 helicase . Thus , save for the single substitution in NS4B tested here , mutations that enhance RNA replication interfere with virus assembly .
Production of infectious HCV in cell culture so far is only possible with the genotype 2a isolate JFH-1 which replicates to very high levels without requiring REMs . In contrast , all genotype 1 isolates described until now replicate very poorly and need enhancing mutations . As shown in this report , at least in the context of Con1 , but probably also for other genotype 1 isolates , with the exception of the K1846T substitution in NS4B REMs interfere with virus production . This is an important finding for two reasons: First , these results support earlier assumptions that REMs augment RNA-replication via different mechanisms [17] . While most of the mutations we analyzed more or less completely abolished virus production , the K1846T mutation in NS4B elevated RNA replication but interfered with virus formation only to a minor extent . These results clearly point to qualitative differences in the mode by which REMs modulate RNA replication and ( in ) -directly virus production . Second , these data clarify why Huh-7 cells transfected with genomes containing REMs [25] , [26] failed to produce virus particles . It is not due to unfavourable host cell conditions like the lack of assembly factors , but rather a consequence of these mutations especially those in NS3 that interfere with virus production , most likely particle assembly . These data therefore explain the attenuation of adapted Con1 genomes in vivo [27] . In fact , in an earlier study we had shown that a Con1 genome containing the three adaptive mutations of the Con1/NS3+S2197P construct ( E1202G and T1280I in the helicase and S2197P in NS5A ) was unable to establish an infection upon intrahepatic inoculation of a chimpanzee . A Con1 genome with only the NS5A mutation ( S2197P ) was attenuated and rapidly reverted to wild type . Taking the data from the present study into account , we can assume that these genomes replicated in RNA-inoculated hepatocytes of the chimpanzee , but due to impaired assembly progeny virus was not produced by Con1/NS3+S2197P and therefore the infection was abortive . Since the Con1/S2197P mutant still releases core protein ( virus ) , albeit to very low levels , initial virus spread in the animal most likely was very limited until the mutant had reverted to wild type . We note that REMs have also been described extensively for the genotype 1a isolate H77 [16] . These mutations reside primarily in the center of NS5A , but cooperative mutations have also been found in the helicase [42] . Although initial attempts to produce infectious H77 virus in cell culture failed with genomes carrying these mutations [42] , a highly adapted genome containing 5 REMs has been described recently that replicates to levels comparable to JFH-1 [43] , [44] . Most notably , cells transfected with this H77-S genome release infectious virus particles , but the amounts are very low . Moreover , the specific infectivity calculated as the ratio of HCV RNA molecules ( genomes ) per infectious unit was about 400-fold lower as compared to JFH-1 ( 5 . 4×10e4 vs . 1 . 4×10e2 , respectively ) [44] . The reason why H77-S transfected cells release such high amounts of HCV RNA is unclear . However , the low buoyant density of the RNA in density gradients and the presence of NS3 and NS5B in these fractions suggest that replication complexes possibly released from dying cells due to cytotoxicity of the efficiently replicating H77-S genome may in part account for these high RNA copy numbers in culture supernatants . Owing to poor replication , infectivity assays of Con1/wt and Con1/K1846T viruses were extremely difficult . Although replication of the wild type genome could be stimulated with the kinase inhibitor H479 , only a very low number of NS5A positive cells became detectable . In case of the Con1/K1846T genome , intrinsic replication efficiency of this genome was still too low for unambiguous detection of viral RNA or proteins . Inclusion of additional REMs either reduced RNA replication ( in case of REMs residing in NS5A ) [17] or blocked core release ( in case of REMs residing in NS3 ) . Furthermore , treatment of Con1/K1846T transfected cells with the kinase inhibitor H479 reduced rather than enhanced RNA replication , comparable to what has been described for REMs residing in NS5A [11] . Finally , attempts to adapt Con1/wt or Con1/K1846T genomes to continuous Huh-7 cell culture failed , because the genomes could not be maintained in passaged cells or culture supernatants , due to insufficient replication capacity ( V . L . and R . B . , unpublished ) . Although the underlying mechanism interfering with virus assembly is unclear , our result argues for a cross-talk between structural and non-structural proteins during the assembly process . In fact , several TEMs have recently been described in the context of JFH-1 and various JFH-1-based infectious chimeras . These mutations reside in the region encoding core to NS2 , but very often in the NS3 helicase domain and the RNA binding replicase factor NS5A [18]–[22] , [45]–[47] . The mutations stimulate production of infectious virus particles without major effects on RNA replication arguing that the viral NS proteins modulate the efficiency of virus production [18] . In fact , we and others have recently shown that NS5A plays a very critical role in the assembly process , which occurs in close proximity of lipid droplets [6] , [7] , [48]–[50] . Core protein accumulates on the surface of these organelles and appears to recruit NS5A or the replicase complex to these sites to trigger virus assembly [48] . It was also found that alterations of NS5A phosphorylation , for which casein kinase I appears to play a major role [10] , [38] , have a strong impact on NS5A – core interaction and virus assembly [7] , [8] and that most , if not all , REMs reduce NS5A hyperphosphorylation [16] , [51] , [52] . Finally , pharmacological inhibition of NS5A hyperphosphorylation enhances RNA replication as is the case with REMs [11] . The current model of HCV assembly that emerges from these observations assumes that via its domain 2 core protein efficiently localizes to lipid droplets [53] whereas NS5A is primarily a component of the replicase complex . We speculate that depending on its phosphorylation status , NS5A is recruited to lipid droplets to interact with the core protein in a way that the viral RNA genome is transferred to core , thus triggering virus assembly . In this respect REMs described here may interfere with the interaction between NS5A and core or recruitment of NS5A to lipid droplets or the RNA transfer from the replicase ( helicase , NS5A ) to the core protein thereby attenuating virus production . The fact that most REMs enhance RNA replication could therefore be due to retention of the viral RNA within the replication complex at the expense of RNA transfer to lipid droplets and/or RNA delivery to the core protein . In this context it is important to note that enhanced replication itself is not responsible for the interference with virus assembly since an adapted Con1 genome with an inactive NS5B polymerase still does not support virus production ( data not shown ) whereas the analogous replication deficient genome lacking REMs does ( Fig . 1D ) . Moreover , JFH-1/wt supports assembly in spite of highly efficient RNA replication . Therefore , we hypothesize that REMs may arrest the viral RNA in a state that prevents the assembly process . The low-level release of virus from H77-S transfected cells may be due to an alternative assembly/release pathway that predominates under these experimental conditions . Clarification of these hypotheses requires more insights into the mechanisms of HCV particle assembly and release . Although extensive tests with other HCV isolates have not been performed we hypothesize that non-adapted consensus genomes , at least those with proven in vivo infectivity , will also support production of infectious virus particles in transfected Huh-7 cells . However , owing to the very low replication levels of these genomes , demonstration of infection of cell cultures will be very difficult , even when stimulating replication e . g . by kinase inhibitors . As shown here , inoculation of xenografted mice with cell culture grown HCV particles is an alternative that is more robust and reliable . In fact , infection of uPA-SCID mice with supernatants of Con1/wt or Con1/K1846T transfected cells resulted in a well detectable viremia . In contrast , supernatants of Con1/NS3+K1846T transfected cells turned out to be non-infectious although these supernatants also contained viral RNA and low amounts of core protein . The nature of these RNA/core structures is not known but due to their low abundance they are not amenable to a biophysical characterization . They may correspond to lipid-containing replication complexes that were released from dying cells , similar to what we and others described earlier [25] , [44] . For several positive strand RNA viruses it has been shown that RNA translation , replication and assembly are tightly coupled [54]–[57] . This coupling may act as a proof-reading mechanism to exclude from progeny particles those viral genomes that have a defect in either translation or RNA replication . As shown here a HCV genome that is unable to replicate ( Con1/D318N; Fig . 1 ) still releases core protein to an amount comparable to the wild type . Although formal proof is missing that this core protein indeed corresponds to virus particles , our data suggest that HCV particle assembly may occur even in the absence of RNA replication . In summary we demonstrate the production of infectious HCV particles in the Huh-7 cell line upon transfection with the genotype 1b isolate Con1 . The interference of REMs with the assembly process provides an explanation why earlier attempts to produce infectious HCV in cell culture were of very limited success . Although this hurdle has in principle been overcome with the identification of the JFH-1 isolate , more replication and assembly competent HCV isolates are urgently needed to cover the full spectrum of genotypes , especially those that are poorly accessible to antiviral therapy . The observation that infectious HCV particles can be produced in Huh-7 cells by the genotype 1b isolate Con1 may provide a new starting point that likely can be extrapolated to other isolates with proven in vivo infectivity .
Huh-7 cell clones Huh7-Lunet [58] and Huh7 . 5 [42] that both are highly permissive for HCV RNA replication were used for electroporation and infection assays . Cells monolayers were grown in Dulbecco's Modified Eagle Medium ( [DMEM] Life Technologies GmbH , Karlsruhe , Germany ) supplemented with 2 mM L-glutamine , nonessential amino acids , 100 U of penicillin per ml , 100 µg of streptomycin per ml , and 10% fetal calf serum ( complete DMEM ) . Cells were routinely subpassaged twice a week at a ratio of 1∶4 to 1∶10 , depending on confluency . For infection experiments , Huh7 . 5 cells were seeded 24 h prior to infection into 12-well plates or on glass cover-slips contained in 24-well plates . Cell densities ranged from 2 to 5×104 cells per well in case of a 12-well plate and 1 to 3×104 cells per well in case of a 24-well plate . All full-length HCV Con1 constructs are based on the consensus clone of the HCV isolate Con1 [59] [AJ238799] . Generation of pFK-Con1/NS3+S2197P , pFK-Con1/NS5A and pFK-Con1/D318N ( Con1/D318N ) has been described recently [27] . Plasmid pFK-Con1/NS3+S2197P differs from pFK-Con1 by 5 nucleotide exchanges ( A3946G , C4180T , C6842T , C6926T , and T6930C ) . Three of these mutations cause amino acid substitutions ( E1202G , T1280I and S2197P ) , whereas the remaining changes are silent . Construct pFK-Con1/NS5A contains a single amino acid substitution ( S2197P ) and two silent nucleotide changes ( C6842T and C6926T ) . The pFK-Con1/D318N plasmid encodes a replication-deficient variant of Con1 that carries a single amino acid substitution changing the GDD motif of the NS5B polymerase to D318N [15] . Constructs pFK-Con1/ΔE1-E2 , pFK-Con1/A358ins and pFK-Con1/NK367AA were generated by PCR-based mutagenesis of pFK-Con1 . PFK-Con1/ΔE1-E2 carries an in frame deletion encompassing amino acid residues 200 to 542 deleting most of the E1 and E2 coding region . Construct pFK-Con1/A358ins encodes a Con1 genome with an insertion of an alanine residue after amino acid 358 which is located in the transmembrane domain of E1 . This mutation was shown to abrogate heterodimerization of E1 and E2 [31] . Variant pFK-Con1/NK367AA comprises two mutations in the transmembrane region of E1 , replacing asparagines 367 and lysine 370 by alanine residues . Similar to A358ins , also this mutant was shown to abrogate heterodimerization of HCV glycoproteins [30] . Constructs pFK-Con1/K1846T and pFK-Con1/NS3+K1846T were generated by insertion of a SfiI-SfiI HCV genome fragment isolated from pFK-I341Luc/NS3-3′/K1846T and pFK-I341Luc/NS3-3′/ET [17] . Constructs pFK-Con1/NK367AA+K1846T and pFK-Con1/A358ins+K1846T were generated by insertion of a SfiI-SfiI HCV genome fragment isolated from pFK-Con1-K1846T . Generation of subgenomic replicons pFK-I389Luc/NS3-3′/wt and pFK-I389Luc/NS3-3′/GND has been already described [15] . Plasmids pFK-I389Luc/NS3-3′/K1846T and pFK-I389Luc/NS3-3′/NS3+K1846T were generated as described above by transfer of a SfiI-SfiI HCV genome fragment . All mutations were verified by DNA sequence analysis . The exact cloning strategies used to generate these constructs can be obtained upon request . PFK-based plasmids were restricted wit AseI and ScaI , whereas puC-based plasmids were linearized with XbaI . Digested plasmids were extracted with phenol and chloroform , ethanol precipitated and dissolved in RNase-free water . In vitro transcription mixtures comprised 80 mM HEPES [pH7 . 5] , 12 mM MgCl2 , 2 mM spermidine , 40 mM dithiothreitol ( DTT ) , each nucleotide triphosphate at a concentration of 3 . 125 mM , 1 U RNasin/µl of reaction volume , 0 . 1 µg restricted plasmid DNA/µl , and 0 . 6 U of T7 RNA polymerase/µl . Reactions were incubated for 2 h at 37°C , an additional 0 . 3 U T7 RNA polymerase/µl was added and the mixture as incubated another 2 h . Transcription was terminated by addition of 1 . 2 U of RNase free DNase ( Promega ) per µg of plasmid DNA and incubation for 30 min at 37°C . After extraction with acidic phenol and chloroform , RNA was precipitated with isopropanol and dissolved in RNase-free water . RNA concentration was determined by measuring absorbance at 260 nm , and the integrity of the transcripts was verified by denaturing formaldehyde agarose gel electrophoresis . Single cell suspensions of Huh-7 , Huh7-Lunet and Huh7 . 5 cells were prepared by trypsinization of monolayers . Detached cells were washed once with PBS and resuspended in cytomix [60] containing 2 mM ATP and 5 mM glutathione at a concentration of 1 . 5×107 cells per ml in case of Huh7 . 5 cells or 1×107 cells per ml in case of Huh-7 and Huh7-Lunet cells . Ten µg of in vitro transcript was mixed with 400 µl of the cell suspension , and electroporated at 960 µF and 270 V by using a Gene Pulser system ( Bio-Rad , Munich , Germany ) and a cuvette with a gap width of 0 . 4 cm ( Bio-Rad ) . Depending on the amount of transfected cells required for the respective experiment either a single electroporation was performed , or cells from several electroporations were pooled and seeded into culture dishes . Total cellular RNA for Northern blots was prepared by a single-step isolation method [61] . For RNA detection by quantitative PCR , the NucleoSpin RNAII kit ( Macherey-Nagel , Düren , Germany ) was employed and used according to the instructions of the manufacturer . For Northern blotting , total cellular RNA was denatured by treatment with 5 . 9% glyoxal in 50% dimethylsulfoxide and 10 mM sodium phosphate buffer [pH 7 . 0] at 50°C for 1 h . Subsequently , RNA was resolved by denaturing agarose gel electrophoresis and transferred to a positively charged nylon membrane ( Hybond-N+; Amersham Pharmacia Biotech , Freiburg , Germany ) with 50 mM NaOH using a vacuum manifold . After drying and crosslinking by UV irradiation , hybridization was performed according to standard protocols [62] . HCV-specific RNA was detected using a [32P]-labeled negative sense riboprobe complementary to NS5B and the 3′ UTR ( nucleotides 8374–9440 ) . HCV-specific bands were quantified by phosphimaging using a BAS 2500 scanner from Fuji . HCV core protein expressed within cells or secreted into the culture medium was quantified using the commercially available Trak C Core ELISA ( Ortho Clinical Diagnostics , Neckargemünd , Germany ) according to the instructions of the manufacturer . When intracellular core expression was determined , cells were lysed in ice cold PBS supplemented with 1% Triton-X-100 , 1 mM PMSF and 0 . 1 µg/ml Aprotinin . Lysates were cleared at 20 , 000×g for 10 min . and supernatants were measured at a dilution of 1∶50 ( or higher ) in PBS . Cell culture medium was filtered through 0 . 45 µm pore size filters and either directly used for ELISA or diluted with PBS prior to measurement . Total RNA prepared from gradient fractions , infected cells or magnetic beads was eluted from NucleoSpin RNAII columns in a volume of 40 µl RNase-free water . Five microliters of the respective sample were used for quantitative RT-PCR analysis employing an ABI PRISM 7000 Sequence Detector ( Taqman; Perkin-Elmer ) . Amplifications were conducted at least in duplicate with the One Step RT-PCR Kit ( Qiagen , Hilden , Germany ) using the following primers and 3′-phosphate-blocked , 6-carboxyfluorescine ( 6-FAM ) - and tetrachloro-6-carboxyfluorescine ( TAMRA ) -labeled probes ( TIB Molbiol , Berlin , Germany ) : HCV-Con1 Taqman probe , 5′-6FAM-TCC TGG AGG CTG CAC GAC ACT CAT-TAMRA-3′; HCV-Con1-S66 , 5′-ACG CAG AAA GCG TCT AGC CAT-3′; and HCV-Con1-A165 , 5′-TAC TCA CCG GTT CCG CAG A-3′; HCV-JFH1 Taqman probe , ; HCV-JFH1-S147 , 5′-TCT GCG GAA CCG GTG AGT A-3′; HCV-JFH1-A221 , 5′-GGG CAT AGA GTG GGT TTA TCC A-3′ . Reactions were carried out in three stages under the following conditions: stage 1 , 60 min at 50°C ( reverse transcription reaction ) ; stage 2 , 15 min at 95°C ( heat inactivation of reverse transcriptase and activation of Taq polymerase ) ; stage 3 , 40 cycles , with 1 cycle consisting of 15 sec at 95°C and 1 min at 60°C . The total reaction volume was 15 µl and contained the following components: 2 . 66 µM 6-carboxy-X-rhodamine ( Rox , passive reference ) , 4 mM MgCl2 , 0 . 66 mM deoxynucleoside triphosphates , 0 . 266 µM probe , 1 µM ( each ) sense and antisense primer , and 0 . 6 µl of enzyme mix . The amounts of HCV RNA were calculated by comparison to serially diluted in vitro transcripts included in the qRT-PCR analysis . About 20 to 30 ml of filtered cell culture medium derived from Huh-7 cells 24 h or 72 h post transfection were concentrated via ultracentrifugation over a 40% or 60% ( wt/vol ) iodixanol ( Optiprep; Invitrogen , Karlsruhe , Germany ) density cushion ( ρ = 1 . 215 g/ml , or 1 . 320 g/ml respectively ) , prepared in CSM ( 0 . 85% [wt/vol] NaCl , 10 mM Tricine-NaOH [pH 7 . 4]; ρ = 1 , 006 g/ml ) , in a SW28 rotor for 7 h at 100 , 000×g ( RCFavg ) at 4°C . Density cushion and interface were resuspended and used for infection or virus capture assays . Alternatively , resuspended material was transferred to the bottom of a fresh tube , and overlaid with a linear iodixanol gradient ( 60% to 0% ) and spun for 18 h in a SW41 rotor at 110 , 000×g ( RCFavg ) at 4°C . Twelve fractions ( 1 ml each ) were harvested from the top . The amount of HCV core protein in 100 µl of each fraction was determined by using Trak C ELISA . For quantifying HCV RNA , 100 µl of the respective gradient fraction were used for RNA preparation with the NucleoSpin RNAII kit ( Macherey-Nagel , Düren , Germany ) . Five µl of the eluate ( equivalent to 12 . 5% of the sample ) were used for quantitative RT-PCR . For the production of lentiviral HCV pseudoparticles ( HCVpp ) , 293T cells were transfected by using the calcium phosphate method essentially as described . Briefly , 2 . 5×106 293T cells were seeded in 10-cm diameter plates 1 day before transfection with 2 . 7 µg of phCMVΔCE1-E2 ( Con1 ) , 8 . 1 µg of HIV-Gag-Pol expression construct [pCMV_R8 . 74 [63]] , and 8 . 1 µg of the lentiviral vector pHR′-CMV-GFP [64] ( where CMV is cytomegalovirus and GFP is green fluorescent protein ) . The medium was replaced 8 h after transfection . Supernatants containing the pseudo-particles were harvested 48 h later , cleared by passage through 0 . 45-µm-pore-size filters . Cleared HCVpp-containing culture fluids were used for immuno-capture assays . The equivalent of 25 µl of Dynabeads Protein A ( Dynal , Invitrogen ) slurry were washed according to the instructions of the manufacturer and coupled to 4 µg of human monoclonal antibody by continuous shaking at room temperature in a total volume of 25 µl sodium phosphate buffer [pH 8 . 1] for 40 min . Beads were washed 3 times with 0 . 5 ml sodium phosphate buffer and blocked 1 h at room temperature with 200 µl PBS , 2% bovine serum albumin . After 3 washes with PBS , beads were incubated with 100 µl filtered supernatant containing HCVpp for 1 h at 4°C by shaking , washed 5 times with PBS and resuspended in a final volume of 100 µl PBS . Bound p24 on the beads was measured by using the INNOTEST HIV Antigen mAb kit ( Innogenetics ) with minor modification of the protocol . In brief , beads were incubated with 100 µl conjugate buffer 1 to release bound p24 and centrifuged 1 min at 14 , 000 rpm . Supernatant was diluted 1∶50 into conjugate buffer 1 and further processed according to the standard protocol of the supplier . Human monoclonal antibodies employed for virus capture assays were described previously [65] . For capture of cell culture derived HCV Con1 particles , 5 µg of purified human monoclonal antibodies ( CBH5 , or the negative control RO4 antibody ) coupled to protein A magnetic beads ( Dynal ) were incubated with filtered and concentrated cell culture medium for 1 h using an overhead rotor at 4°C . Beads were washed 5 times with 1 ml PBS , and RNA was extracted by using the NucleoSpin RNAII kit ( Macherey-Nagel ) . Captured RNA was quantified by using TaqMan qRT-PCR . Virus particles contained in culture supernatant were concentrated as described above . For each capture assay 5 µg of CBH5 or RO4 antibody was covalently coupled to tosyl-activated Dynabeads ( Invitrogen , Germany ) according to the instructions of the manufacturer . Virus particles were captured as described above , beads were washed 5-times with 1ml of PBS supplemented with 1 mM CaCl2 and resuspended in 100 µl of the same buffer . Immune complexes were left untreated or incubated with 0 . 5% Triton X-100 for 5 min at RT in the presence or absence of 2 . 5 µl RNAsin ( Promega , Mannhein , Germany ) . Thereafter , beads were incubated in the absence or presence of 2U S7 nuclease ( Roche Mannheim , Germany ) for 30 min at 37°C and the nuclease was inactivated by the addition of 2 mM EDTA . RNA was extracted by using the NucleoSpin RNAII kit ( Macherey-Nagel , Düren , Germany ) and RNA was quantified by using TaqMan qRT-PCR . Transient HCV RNA replication assays were performed as described previously above . Huh7 . 5 cells were electroporated with a subgenomic Con1/wt or Con1/K1846T or Con1/D318N or Con1/NS3+K1846T luciferase replicon and resuspended in 20 ml culture medium . Aliquots of 2 ml each were seeded per well of a 6-well plate and replication was determined by measuring luciferase activity 4 , 24 , and 48 h post-transfection . Four hours after electroporation , medium was removed and cells were treated with various concentrations of H479 . Luciferase activities were normalized to the DMSO control value to determine the fold induction of replication . Statistical analyses were conducted using PRISM4 ( GraphPad Software Inc . ) and two-way ANOVA test . Supernatants from 10 electroporations of Huh7 . 5 cells with each 10 µg Con1/wt or Con1/K1846T or Con1/NS3+K1846T were harvested 12 and 24 h post electroporation and pools were filtered through 0 . 45 µm-pore-size filters . Filtrates were loaded onto a cushion composed of 4 ml 10% Optiprep-PBS and 3 ml 28% Optiprep diluted in serum-free-medium . Samples were centrifuged for 4 h at 100 , 000×g in a SW28 rotor ( Beckman ) at 4°C . Virus concentrated at the interface of the cushion was recovered and concentrated by centrifugation using a Centricon Plus-70 centrifugal filter device ( 100K NMWL; Millipore , Germany ) according to the instructions of the manufacturer . Concentrated virus was resuspended in 1 ml culture medium . Based on core-ELISA measurements we calculated that about 50% of core protein was recovered in this final preparation and that core protein was concentrated 200-fold . Huh7 . 5 cells seeded on glass cover slips were pretreated with or without 25 nM of Concanamycin A for 1 h at 37°C; thereafter cells were infected with 300 µl of the concentrate in the presence or absence of 25 nM Concanamycin A ( Sigma ) . Four hours later cells were washed with fresh medium and incubated in complete medium with or without H479 ( 10 µM ) . After 72 h , cells were fixed with icecold methanol . Immunolabelling of NS5A was performed with a monoclonal antibody specific for NS5A ( Virostat , Portland , USA ) at a dilution of 1∶50 in PBS supplemented with 5% normal goat serum . NS3 was detected with a rabbit polyclonal antiserum obtained by immunization with a recombinant NS3 fragment of JFH-1 ( amino acid residues 293 to 631 ) . Bound primary antibodies were detected by using a goat antibody conjugated to AlexaFluor488 ( Invitrogen , Germany ) at a dilution of 1∶1 , 000 in PBS/5% normal goat serum . Nuclear DNA was counterstained with DAPI ( Molecular Probes , Eugene , OR ) . Images were acquired with an inverted fluorescence microscope ( Leica , Germany ) . The mouse study was conducted at the Ghent University Hospital , with protocols approved by the Ethical Committee and Animal Ethics Committee of the Ghent University Faculty of Medicine . Transgenic SCID mice overexpressing the uPA gene under the control of an albumin promoter ( uPA+/+-SCID ) were xenografted with primary human hepatocytes as described elsewhere [66] . Chimeric mice were inoculated by intraperitoneal injection of 100 µl of purified and concentrated culture supernatant ( prepared as described above ) . Inocula contained 2×108 IU HCV RNA for each preparation and 540 pg core protein for Con1/wt , 200 pg for Con1/K1846T and 23 pg core protein for Con1/NS3+K1846T . EDTA plasma samples were collected at weekly intervals after inoculation and infection was monitored by a commercial qRT-PCR kit ( Roche COBAS AmpliPrep/TaqMan48 assay , Roche Diagnostics ) . Due to dilution of the samples , the detection limit of the test was 750 IU/ml . HCV RNA was isolated from 110 µl of serum taken from mouse K831 six weeks post inoculation with Con1/K1846T . Viral RNA was isolated by using the Nucleo Spin RNA Virus Kit ( Macherey-Nagel , Germany ) as recommended by the manufacturer . RT-PCR was performed with the Expand-RT system ( Roche , Germany ) according to the instructions of the manufacturer using primer A-9413 ( 5′-CAG GAT GGC CTA TTG GCC TGG AG-3′ ) for cDNA synthesis . First amplification was done by using the Expand Long Template PCR Kit ( Roche ) and primers S-4542 ( 5′-GAT GAG CTC GCC GCG AAG CTG TCC-3′ ) and A-6156 ( 5′-CGC TCT CAG GCA CAT AGT GCG TGG-3′ ) . Because of low RNA levels nested PCR was required , for which we used primers S-4542 and A-6103 ( 5′-GCT ATC AGC CGG TTC ATC CAC TGC-3′ ) . Amplified fragment was inserted into pFK-I389Luc-EI/NS3-3′/JFH1-dg after restriction with NsiI and HpaI . Sequence analysis was performed using primer A-8242 ( 5′- CGT TGG GCA GGG GAG TAC TGG AAG -3′ ) . | The hepatitis C virus ( HCV ) is a major cause of acute and chronic liver disease . Unusual for a positive strand RNA virus , HCV has the high propensity to establish persistent infection , which increases the risk for liver cirrhosis and hepatocellular carcinoma . No selective therapy is available thus far and its development has been hampered by the lack of adequate cell culture systems . With the advent of subgenomic replicons , i . e . RNAs containing only the viral replicase genes and that self-amplify in the human liver cell line Huh-7 , this hurdle has been overcome to some extent . However , save for a single genotype 2a isolate , efficient replication of all HCV isolates described thus far requires replication enhancing mutations ( REMs ) , but genomes with REMs do not support production of infectious virus particles . In this study we show that except for one mutation in non-structural protein 4B , REMs interfere with the assembly of infectious virus particles , whereas an unaltered HCV genome supports production of cell culture–derived virus that is infectious in vitro and in vivo . Our observations provide an explanation for the attenuation of cell culture adapted HCV genomes and open new perspectives for the development of culture systems for difficult to treat HCV genotypes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology"
] | 2009 | Production of Infectious Genotype 1b Virus Particles in Cell Culture and Impairment by Replication Enhancing Mutations |
Microtubules are long filamentous hollow cylinders whose surfaces form lattice structures of αβ-tubulin heterodimers . They perform multiple physiological roles in eukaryotic cells and are targets for therapeutic interventions . In our study , we carried out all-atom molecular dynamics simulations for arbitrarily long microtubules that have either GDP or GTP molecules in the E-site of β-tubulin . A detailed energy balance of the MM/GBSA inter-dimer interaction energy per residue contributing to the overall lateral and longitudinal structural stability was performed . The obtained results identified the key residues and tubulin domains according to their energetic contributions . They also identified the molecular forces that drive microtubule disassembly . At the tip of the plus end of the microtubule , the uneven distribution of longitudinal interaction energies within a protofilament generates a torque that bends tubulin outwardly with respect to the cylinder's axis causing disassembly . In the presence of GTP , this torque is opposed by lateral interactions that prevent outward curling , thus stabilizing the whole microtubule . Once GTP hydrolysis reaches the tip of the microtubule ( lateral cap ) , lateral interactions become much weaker , allowing tubulin dimers to bend outwards , causing disassembly . The role of magnesium in the process of outward curling has also been demonstrated . This study also showed that the microtubule seam is the most energetically labile inter-dimer interface and could serve as a trigger point for disassembly . Based on a detailed balance of the energetic contributions per amino acid residue in the microtubule , numerous other analyses could be performed to give additional insights into the properties of microtubule dynamic instability .
Microtubules ( MTs ) are cellular organelles that participate in major cellular processes such as mitosis , cell shape maintenance , cell motility and motor protein transport and constitute a major target for a wide range of drugs , most notably anti-mitotic chemotherapy agents such as paclitaxel . Due to their importance in cell biology , MTs have been the topic of active research into their structure and function for several decades [1] . The pivotal role of MTs in cell division , by forming the mitotic spindle that segregates chromosomes , makes them an important target for antimitotic cancer chemotherapy drugs [2 , 3] . The peanut-shaped αβ-tubulin heterodimer is the building block of MTs [4] . Tubulin heterodimers associate longitudinally to form protofilaments , which in turn associate laterally to form a left-handed three-start helix with a seam , that results in the most common microtubule structure , the so-called B lattice [5] . Since tubulin dimers polymerize end to end , MTs become polarized , meaning that one end has α-subunits exposed ( minus end ) while the other end where faster growth usually occurs has β-subunits exposed ( plus end ) ( Fig 1A ) [6] . Within a tubulin heterodimer , GTP binds at the α-tubulin N-site which occurs at the intra-dimer interface . This GTP molecule does not undergo hydrolysis . Another GTP molecule attaches at the β-tubulin E-site and undergoes hydrolysis to GDP and phosphate shortly after assembly [7] , in a process which drives the stochastic switching between growth and shrinkage in MTs . This unique property of microtubules is commonly referred to as dynamic instability [8] . Mitchison and Kirschner proposed the so-called GTP-cap model , which states that as long as the plus end of an MT is capped with GTP , it continues to grow . However , if GTP hydrolysis is sufficiently fast to catch up to the growing tip of the MT , rapid shrinkage , called a catastrophe , results [9] . Upon binding to an MT , some pharmacological agents such as taxol or epothilone stabilize the system and inhibit shrinkage [10] . Several studies have been conducted to determine which specific structural transitions that accompany GTP hydrolysis or taxol binding are responsible for their effect on MT stability , especially the transition of the tubulin dimer between its straight and curved states [11–15] . In the most recent of these studies , Alushin et al . found that GTP hydrolysis leads to a compaction around the E-site nucleotide which is reversed upon taxol binding [15] . This compaction was proposed to generate a strain which is powered by the energy of GTP hydrolysis and is believed to be released only through outward curving of protofilaments , initiating disassembly [16] . A missing component in these studies , however , is the quantification of the free energy changes that accompany these structural transitions . Due to the difficulties related to its experimental measurements , many simulations have been conducted to study detailed MT energetics [17–22] . In a recent study we have analyzed the strength of hydrogen bonds that bring and hold tubulin subunits together within different lattice configurations [23] . However , in all of these simulations , several factors were still missing . Most importantly , the full energetics of a complete MT model , which is essential to understanding the thermodynamics of tubulin assembly , has not been estimated yet due to the high computational price associated with such analyses . A detailed energy balance involving contributions due to each residue , domain or subunit , to the best of our knowledge , was never considered . As a result of recent advances in computational technology , GPU-based computations can now be implemented to perform very demanding calculations in a reasonable amount of time . With this technology readily available , we simulated two complete all-atom MT models and studied in detail their energetics . The models studied are: ( a ) an MT with GDP in the E-site ( GDP-Model ) and ( b ) an MT with GTP in the E-site ( GTP-Model ) . We did not need to look for a non-hydrolyzable analogue of GTP as hydrolysis is not a problem in molecular dynamics simulations , in contrast to experimental procedures [16] . The MT model that we used was initially built by Wells and Aksimentiev [24] utilizing sophisticated theoretical techniques to combine experimental structural information from a cryo-electron microscopy map of MT at 8 Å resolution [25] and electron crystallography structure of tubulin at 3 . 5 Å resolution [26] . We combined this model with the recently published crystal structures [15] in order to generate an atomistic representation involving an infinite number of infinitely long MTs . This is possible due to the use of periodic boundary conditions . ( see S1 Movie ) .
A 50-ns MD trajectory was analyzed for several equilibration aspects , the first of which is the root-mean-square deviation ( RMSD ) of the backbone atoms relative to the starting structure . In addition , two nearly perpendicular MT cylinder diameters , namely Dx and Dy , were also calculated along the trajectory . Referring to the tubulin dimer numbering in Fig 1B , the diameter Dx was defined as the distance between the center of mass of dimer 4 and the center of mass of dimer 10 and 11 , while Dy was defined as the distance between the center of mass of dimer 1 and the center of mass of dimer 7 and 8 . In both diameters , only the distance projection on the x-y plane was considered as this is what gives the cylinder diameter . Plots showing the change in RMSD of the backbone atoms , Dx and Dy over simulation time for the GDP- and GTP-Models are shown in Fig 2A and 2B . The two diagrams indicate a strong correlation between fluctuations in RMSD and in diameters which indicates that most of RMSD fluctuations are due to changes in the circular shape of MT cylinders rather than the rearrangement of domains . The two diagrams also show the flexibility of MT cylinders as they deform spontaneously from a circular to an oval shape and vice versa . Movies showing the change of the two diameters over simulation time can be found in Supporting Information ( see S2 and S3 Movies ) . Since our particular interest is in the MT energetics , we used the overall MT energy across lateral and longitudinal inter-dimer interfaces as an indication of whether the system is equilibrated or not . Hence , we calculated these energies using MM/GBSA and the formula in Eqs 1 and 2 and plotted the total energy per MT ring versus simulation time ( Fig 2C and 2D ) . Both plots indicate that the overall lateral and longitudinal energies in both the GDP- and GTP-Models have already equilibrated at least before the last 20 ns of the MD simulation time . The plots also show that the large fluctuations in RMSD or Dx and Dy hardly affect the MT energetics at either of the two interfaces , which is a good indication of the energetic stability of our models . Total breakdown of the predicted energy contributions enabled us to perform the analysis for different residues , domains , subunits , and dimers across both lateral and longitudinal inter-dimer interfaces . Before listing the results , it should be noted that energies calculated via the MM/GBSA method do not necessarily reflect absolute energy values . Rather , they are used for relative comparison within the same model [27] . It should also be noted that all energies listed here are calculated per MT ring , unless otherwise specified . As Table 1 summarizes , the overall energy of interaction across the 13 lateral tubulin interfaces ( see Fig 1B ) , E t o t l a t , was found to be −411±29 kcal/mol , nearly 60% of which is due to α-α interactions and the rest is due to β-β interactions . On the other hand , the contribution of the dimer acting as a receptor ( see the explanation of the ligand/receptor convention in the methods section and in Fig 3A and 3B ) , E R l a t , was about 54% of the overall energy while the rest was attributed to the ligand , E L l a t , with the difference entirely attributed to solvation effects rather than direct interactions . It should be noted , however , that the α subunit of the ligand ( Lα ) and the β subunit of the receptor ( Rβ ) together contribute −312±29 kcal/mol which is nearly 75% of E t o t l a t , with the Lα contribution slightly larger than that due to Rβ . The contribution of Lβ and Rα was found to be much smaller , only 25% of E t o t l a t . Upon structural inspection , this 50% difference , being almost entirely due to electrostatic interactions , was attributed to diagonal interactions between subunits; although the interface between Lα and Rβ is dominated by oppositely-charged residues and thus stabilizing the interaction , the opposite is true at the destabilizing interface between Rα and Lβ which has , for example , residues Rα/Glu220 and Lβ/Asp130 destabilizing the lateral interface by 12±1 and 10±2 kcal/mol , respectively . As to the energetic breakdown according to interaction types , the contribution of the van der Waals and non-polar solvation energy , E ( vdW+SA ) , to the overall energy is largely stabilizing with an average value of −1476 kcal/mol , 85% of which is due to the vdW interactions . This stabilization is opposed by destabilization due to electrostatic interactions; the average sum of electrostatic and the polar solvation energy , E ( ele+GB ) , is 1065 kcal/mol . This is expected since tubulin dimers are highly negatively charged and tend to repel each other . Regarding the detailed energy contributions per individual residues , the most important residue across the lateral interface was found to be Rβ/Tyr283 followed by Rα/His283 and Lα/His88 , with overall stabilization energies of −90±5 , −47±5 and −42±3 kcal/mol per MT ring , respectively . Rβ/Tyr283 alone supplies more than 20% of lateral stability most of which is due to the vdW interactions . In fact , most of the stabilizing residues on top of our list were neutral ones with a strong stabilizing vdW component . On the other hand , almost all of the destabilizing residues were charged ones with a strong electrostatic component , most destabilizing of which is Lβ/Lys124 with an energy of 22±7 kcal/mol . A complete list of the different energetic contributions of each residue in the ligand and receptor per MT ring is provided in the Supporting Information . Domain contributions to the overall energy per MT ring were also calculated and Fig 4A and 4B show the most relevant of them . The contribution of the M-loop in both α and β subunits is by far the largest , with values of −112±10 and −159±10 kcal/mol , respectively , making up about two thirds of the energy of the overall lateral interactions . This agrees well with previous predictions , although precise values of their energetic contributions were never calculated [25 , 28 , 29] . Other less important domains are the Lα/N-terminal loop , Lα/H2-S3 loop , Lα/H3 helix and Lα/H9 helix at the α interface with a stabilization of −72±6 , −62±6 , −57±10 and −16±7 kcal/mol , respectively [25 , 28] . Lβ/H3 helix at the β-β interface , however , has a strongly destabilizing effect of 37±8 kcal/mol . This supports previous predictions based on structural analysis by Li et al . and Nogales et al; however , these authors did not specify if these interactions are stabilizing or not [25 , 28] . Additionally , Lβ/H2” helix and Lβ/H1’-S2 loop also have relatively strong stabilizing contributions of −53±7 and −43±5 kcal/mol , respectively . As Table 1 summarizes , the overall interaction energy across the lateral interface in the GTP-Model , E t o t l a t , was found to be −482±29 kcal/mol , nearly 60% of which is due to the α-α interactions . This average overall energy is 71 kcal/mol ( nearly 20% ) more stable than the overall energy of the GDP-Model which explains the role of GTP in stabilizing MTs as will be shown later . Nearly 90% of this difference in stability is solely attributed to enhancement of the contribution of the ligand , both α- and β-subunits , rather than the receptor . As was noticed in the GDP-Model , Lα and Rβ are also responsible for most of the lateral stabilization in the GTP-Model , −338±22 kcal/mol ( 70% of E t o t l a t ) . Upon breakdown of the interaction energy to its individual components , we find that in the GTP-Model , the E ( vdW+SA ) contribution becomes −1432 kcal/mol while E ( ele+GB ) becomes 950 kcal/mol . Comparing this to the GDP-Model , it turns out that GTP destabilizes the vdW and non-polar solvation interactions by 44 kcal/mol and stabilizes electrostatic and polar solvation interactions by 115 kcal/mol , which results in the net stabilization of 71 kcal/mol as mentioned earlier . This difference becomes clear by analyzing Fig 4A and 4B for domain contributions and Fig 5 for residual contributions . It is apparent from Fig 4A that GTP strengthens the contributions of the Lα/H3 helix and Rα/H9 helix by 23±10 and 20±16 kcal/mol , respectively . Most of this helix stabilization can be attributed to interactions involving Rα/Glu290 ( residue number in Fig 5 , i , is 290 ) , residue Lα/Asp127 ( i = 998 ) , and residue Lα/Arg123 ( i = 994 ) . These three residues stabilize the GTP-Model over the GDP-Model by energy values of 31 , 20 and 19 kcal/mol , respectively , mostly due to electrostatic interactions . Upon structural analysis it is apparent that GTP slightly rotates the dimer acting as a ligand toward the one acting as a receptor , thus allowing stronger interactions between H3 and H9 helices with oppositely-charged residues . GTP also enhances the stability imparted by the Lα/H2-S3 loop and the Rα/H10-S9 loop , although it moderately decreases the role of the Lα/N-terminal loop as well as the Rα/M-loop in the overall MT stability . Similar conclusions are reached in regard to the β-subunit and the effect of the Lβ/H2” helix through residue Lβ/Asp90 ( i = 1401 ) and the Rβ/M-loop through residue Rβ/Arg284 ( i = 724 ) . Both domains are stabilized in the GTP-Model by extra 18±10 and 10±15 kcal/mol compared to the GDP-Model , respectively . The charged nature of all these residues explains why most of GTP stabilization is manifested in E ( ele+GB ) not E ( vdW+SA ) . Fig 4B also shows that GTP reduces the destabilization caused by the Lβ/H3 helix and the Lβ/H3-S4 loop . On the other hand , GTP reduces stability imparted by the Lβ/H1’-S2 loop and the Rβ/H9 helix . Details of the contribution of each residue in the GTP-Model can be found in the Supporting Information . Analysis of the strength of interactions across the longitudinal inter-dimer interface in the GDP-Model yielded , as summarized in Table 2 , an overall energy of −1240±32 kcal/mol per MT ring , which is nearly three times the lateral interaction energy . This is in agreement with structural observations [28] . Due to the orientation of tubulin dimers at the longitudinal inter-dimer interface , the contributions of Lα and Rβ are essentially zero and will always be neglected here . On the other hand , the contribution of Lβ is 54% of the total value , and the remainder is contributed by Rα . The breakdown of this energy yields an average E ( vdW+SA ) of −2668 kcal/mol which is almost twice as large as the value across the lateral interface . This is obviously due to the tighter packing of the residues here as opposed to looser packing at the lateral interface . The average E ( ele+GB ) across the longitudinal inter-dimer interface is 1428 kcal/mol and it is 34% larger than its value at the lateral interface . Per-residue energy analysis reveals the most important residues to longitudinal stability , the first of which is Lβ/Arg401 from the H11-H11’ loop which alone supplies −101±7 kcal/mol ( nearly 10% ) [23] . After that come residues Lβ/Phe404 and Lβ/Trp407 from the H11’ helix both of which support longitudinal stability by contributing −91±3 and −78±3 kcal/mol , respectively . This makes the two former domains , which constitute part of the tubulin C-terminal domain , the most critical for longitudinal stability in the β-subunit ( Fig 4D ) . The figure also shows that the following domains in the Lβ subunit: the T5 loop , T3 loop , and T2 loop are also very important for longitudinal stability . The role of the GDP cofactor appears quite influential at the longitudinal inter-dimer interface , in contrast to the lateral one . It is primarily destabilizing with a large contribution of 79±11 kcal/mol due mainly to a strong electrostatic repulsion with the highly negative environment , despite its strong salt bridge with Rα/Lys352 . Residual analysis of the Rα subunit also shows some relatively less important residues; Rα/Trp346 , Rα/Tyr262 and Rα/Lys352 with energy contributions of nearly −60 kcal/mol for each of them . These and other residues are responsible for the following domains in the Rα subunit: the H10-S9 loop , H8-S7 loop , and the S9 strand being the top stabilizers in Fig 4C . The Rα/H8 and Rα/H10 helices are also relatively important for longitudinal stability . Both the Rα/C- and Rα/N-terminal domains are important as well , with the Rα/N-terminal loop being a destabilizer , in contrast to its role at the lateral interface . As summarized in Table 2 , the overall interaction energy across the longitudinal inter-dimer interface in the GTP-Model was found to be −1098±30 kcal/mol per MT ring , which is 141 kcal/mol ( 10% ) less stable than the GDP-Model system . This difference is attributed to a 7% decrease in the Rα and 3% decrease in the Lβ interactions . Upon energetic breakdown we see that GTP destabilizes the vdW and non-polar solvation energy by nearly 250 kcal/mol , while stabilizing electrostatic and polar solvation energy by nearly 110 kcal/mol . This could be due to the longstanding observation that GTP leads to an expansion in the E-site and lengthening of the tubulin dimers . That is , axial dimer repeat changes from 81 . 20 Å in GDP-tubulin to 83 . 38 Å in GTP-tubulin [12 , 15] . This reduces the packing of atoms at the interface and hence lowers both the vdW attraction and electrostatic repulsion , the former being affected most due to its stronger dependence on distance . Looking into domain contributions in Fig 4C and 4D we see how GTP destabilization of longitudinal interactions can be subdivided . The most pronounced difference between the GDP- and the GTP-Model appears in regard to the cofactors at the E-site . Although GDP was largely destabilizing in the GDP-Model , GTP becomes relatively largely stabilizing , with an energy change from the GDP-Model of nearly −125±14 kcal/mol . However , this change should not be considered without taking into account the effect of the Mg2+ ion that accompanies GTP . This magnesium ion introduces an instability of 95±4 kcal/mol to the GTP-Model . Hence , the overall effect of replacing GDP by GTP and a magnesium ion is a stabilization of 30 kcal/mol on average . Other causes of the lack of stability in the GTP-Model Lβ include the decrease in the contribution of the H11’ helix because GTP offsets interactions by Lβ/His406 ( i = 1709 ) by as much as 25 kcal/mol . This is because this histidine is protonated in the GTP-Model and neutral in the GDP-Model and therefore behaves differently in both cases . Being charged in the GTP-Model , it is distracted from the strong attractive vdW interactions it makes with the Rα/H8-S7 loop by electrostatic and hydrogen bonds with other residues within the Lβ subunit . GTP also causes longitudinal stabilization due to the domains: the H2 helix and the T2 loop to decline while causing stabilization due to the H11-H11’ loop and the T5 loop to rise . As to the Rα-subunit ( Fig 4C ) , stabilization due to several domains declines in the GTP-Model . These domains include the T7 loop , the S9 strand , the C-terminal loop , the H10 helix , the H12 helix , the H8-S7 loop , and the H10-S9 loop . In short , the GTP-Model is longitudinally less stable than the GDP-Model in most of the domains occurring at the longitudinal inter-dimer interface . An exception to this rule is the increased stabilization due to the C-terminal tail , the N-terminus and the H8 helix , Fig 4C shows the extent of stabilization or destabilization imparted by GTP on each domain . We should also mention that the strong attraction of the Rα/T7 loop emerging after GTP hydrolysis ( Fig 4C ) could explain the proposed compaction of the E-site after GTP hydrolysis [15] . In fact , the overall increase in longitudinal dimer-dimer attraction after GTP hydrolysis , as shown by the different values of E t o t l o n g in both models , explains the driving force for this E-site compaction . Among other important residues , Rα/Lys352 ( i = 352 ) of the domain S9 strand has a largely reduced contribution in the GTP-Model , as shown in Fig 5 , which is 37 kcal/mol less stabilizing than in the GDP-Model . While having comparable vdW contributions in the two models , this residue suffers strong repulsion probably due to the nearby Mg2+ ion in the GTP-Model . Another important residue is Rα/Val440 , located in the C-terminus of the α-subunit in our model . GTP enhances the stabilization caused by this residue by nearly 33 kcal/mol over the GDP-Model . Additional important residues and their contributions are shown in the Supporting Information . Depolymerizing MTs display protofilaments that peel into “ram’s horns” formations under high magnesium buffer conditions . The ends of MTs become frayed , however , under physiological concentrations of magnesium [11] . The energy profile throughout the longitudinal inter-dimer interface provides a clear explanation for the disassembly mechanism , its driving force , and its relation to Mg2+ concentration . We characterized each residue in the longitudinal subsystems by its radial distance from the MT lumen in Å , which was plotted on the x-axis . The interaction energies of residues , per MT ring , over half-closed intervals of [x , x+3 ) were summed up and plotted on the y-axis to produce the radial energy profiles in Fig 6A , 6B and 6C . The diagram in Fig 6A leads to a striking observation that the energy distribution throughout the longitudinal inter-dimer interface is not even , with the outward portion ( x > 30 Å ) largely outweighing the inward portion ( x < 30 Å ) , with the center of mass of tubulin being at x ≈ 30 Å . To mention specific values , in the GTP-Model , the outward portion provides nearly −956 kcal/mol while in the GDP-Model it provides −982 kcal/mol , both values being larger than 80% of the overall longitudinal interaction energy . This uneven distribution of energy , or forces of attraction , is proposed to yield a strong torque that tends to curl MT protofilaments outwardly , breaking lateral bonds and promoting disassembly as illustrated in Fig 7A . Radial energy profiles of different components of the interaction energy are also shown in Fig 6B and 6C , where electrostatic interactions cause very strong repulsion through the inward portion and attraction only at the periphery where the H11-H11’ loop and particularly residue Lβ/Arg401 are located . We propose a pivotal role for this residue , and for the entire C-terminal domain , in regulating dynamic instability . Electrostatic repulsion by the inner domains and attraction by the outer C-terminal domain is the recipe for outward curling and disassembly in MTs . The vdW distribution will also work , as shown in Fig 6B for the GDP-Model and Fig 6C for the GTP-Model , to curl protofilaments outward until the vdW contacts , and other components , are balanced out . The largely destabilizing Mg2+ ion ( see Fig 4D ) also plays an important role . Even though GDP at the E-site has low affinity for Mg2+ [30] , it may still attract Mg2+ if it is present in high concentrations or Mg2+ may stay in the E-site after GTP hydrolysis . This largely destabilizes the inner portion of the protofilament ( blue dashed arrow in Fig 6A ) , allowing outward forces to pull tubulin out with even less resistance from the other side , thus promoting outward curling and MT disassembly . This explains why large Mg2+ concentrations promote ram’s horns formations [31] and increase the rate of disassembly [32 , 33] , while its low concentrations produce frayed ends and lower rates of disassembly [11] . To explain MT disassembly from a free energy perspective , Fig 7A shows an illustration of the analyzed situation . As already established , uneven distribution of attractive interactions along the longitudinal inter-dimer interface favors outward curling . In the GTP-Model , outward curling is favored by −956 kcal/mol of interaction energy outwardly with respect to the center of mass of tubulin , as compared to −982 kcal/mol in the GDP-Model . These curl-favoring energies/forces are opposed by the lateral interaction energies which tend to pull protofilaments back from both sides , i . e . double the effect . The magnitude of this effect is 2 × E t o t l a t , giving −964 kcal/mol in the GTP-Model which is much larger than −822 kcal/mol in the GDP-Model , all energies given per MT ring . We propose that this lateral inward pull balances out the longitudinal outward push in case of the GTP-Model . That being said , the presence of a GTP cap at the tip of the MT would prevent outward curling and thus provide stability for the entire MT structure . After GTP hydrolysis reaches the cap , however , lateral bonds become weaker and longitudinal outward push manages to break the lateral contacts , causing outward curling and MT disassembly . High concentrations of Mg2+ may also increase outward curling and the disassembly rate , as explained earlier . Similar observation could be made about the tangential energy profiles at the longitudinal inter-dimer interface . Fig 6D , 6E and 6F show the tangential energy profiles with the x-axis showing the distance from the laterally adjacent protofilament . On the x-axis , x < 30 is the tubulin intermediate domain while x > 30 is the nucleotide binding domain with x ≈ 30 being at the center of mass ( see Fig 7B ) . Fig 6D shows that in The GTP-Model , the distribution is also uneven with right-side portion being −1023 kcal/mol ( nearly 93% of the total ) as compared to −887 kcal/mol ( 71% of the total ) in the GDP-Model . This means that in the GTP-Model , there is a strong force tilting it sideways . However , after GTP hydrolysis and rearrangement of domains at the longitudinal inter-dimer interface , that force largely decreases and the uneven distribution starts to balance out , as shown in Fig 6D , decreasing the strain on lattice integrity . This is in perfect agreement with the recent findings of Alushin et al . [15] They observed that GTP hydrolysis and the release of an inorganic phosphate group leaves a hole within the longitudinal inter-dimer interface between tubulin dimers producing a strain which results in sideway tilting in the same direction [15 , 16] . In the present work we show that this tilting is also driven by the uneven energy distribution along the same direction as in the work of Alushin et al . [15] ( see Fig 7B ) . However , this sideway tilting should not be considered as the the driving force for disassembly since it is orthogonal to the outward curling . Combining the two effects together , we conclude that uneven distribution at the longitudinal inter-dimer interface generally leads to a large outward and slight sideway tilting of protofilaments , the former of which is responsible for disassembly of GDP-bound MTs . As mentioned in the Methods section , the MT ring was divided into 13 subsystems of laterally adjacent tubulin dimers and another 13 subsystems of longitudinally adjacent tubulin dimers ( see Fig 3 ) . All of the energies presented earlier were expressed per MT ring , meaning that they were summed over the 13 subsystems . In this section , however , we focus on the interaction energy in each subsystem . Fig 8A and 8B show energy diagrams for lateral and longitudinal interactions superposed over the MT ring . We first note that the shape of the lateral interactions ( Fig 8A ) in the GDP-Model is very distorted with several “kinks” of very low energy . When compared to the GTP-Model , its shape is much less distorted . This could come as a straightforward consequence of the fact that GTP-Model is laterally more stable than the GDP-Model and hence suffers less “deformations” . It is worth mentioning that the deepest of the kinks in the GDP-Model energy diagram , i . e . the interface with the weakest binding energy , is the one occurring at the seam ( between dimer 13 and dimer 1 ) , in contrast to its strength in the GTP-Model . It has a binding energy of −9±7 kcal/mol which is very low compared to the one at the interface between dimer 12 and 13 , for example , which has an energy of interaction equal to −57±9 kcal/mol . We predict that protofilaments number 1 and 13 having very strong longitudinal contacts antagonized by very weak lateral contacts at the seam , will be the first to dissociate laterally and curve outwards . This should open the MT cylinder which should then trigger disassembly . Therefore , MT energetics suggest that the seam is the most labile inter-dimer interface in the MT structure and could act as a trigger point for disassembly . This is precisely what was reported recently [34] . The energy diagrams at the longitudinal inter-dimer interfaces ( Fig 8B ) appear to be more even than at the lateral interfaces . However , we see no major difference in the pattern between the GTP-Model and the GDP-Model except that longitudinal interactions in the GDP-Model are stronger , which was established earlier .
We used sophisticated all-atom molecular dynamics simulations to produce accurate MT models , combined with high resolution cryo-electron microscopy maps , to generate an infinite number of infinitely long MT representations . The MM/GBSA energy analysis that followed the simulations enabled an estimate of the contributions of individual residues , domains , subunits and dimers toward the lateral and longitudinal stability of a complete MT ring . We found that longitudinal interactions are about two to three times stronger than lateral interactions explaining the greater stability of the MT structure along its axis than radially . This finding agrees with previous structural observations [28] and computational estimations [18 , 22] . We also found that interactions are not evenly distributed radially along the longitudinal inter-dimer interface . That is , attractive interactions are largely concentrated away from the MT lumen , producing a force that curls protofilaments outward and eventually causing MT disassembly . The GTP-Model was laterally more stable than the GDP-Model and the opposite was true for the longitudinal inter-dimer interface . Since lateral forces oppose outward curling while longitudinal forces support it , we expect the GTP-Model to be less prone to disassembly than the GDP-Model . With its lateral forces being strong enough to prevent outward curling caused by longitudinal forces , the GTP-cap at the plus end can stabilize an entire MT cylinder . After GTP hydrolysis reaches the cap , lateral forces are too weak to prevent outward curling , especially at the seam which has the weakest lateral contacts . This results in outward curling and microtubule disassembly . We also confirmed that the MT seam is most likely to act as a trigger point for MT disassembly by being the most labile interface in the MT cylinder [34] . Magnesium ion was demonstrated to be an influential factor in MT stability . Being present at the inner portion of the longitudinal inter-dimer interface , the largely destabilizing Mg2+ ion repels the inward portion and enhances outward curling , the formation of ram’s horns structures and rapid disassembly , which is consistent with key experimental findings [11] . This action of Mg2+ at the E-site of tubulin is suppressed by GTP in GTP-capped MTs . As we showed earlier , the ensemble of Mg2+ and GTP at the E-site is collectively stabilizing . However , hydrolysis of GTP and release of inorganic phosphate create a gap at the longitudinal inter-dimer interface and leave the largely destabilizing ensemble of GDP and Mg2+ which rapidly promotes outward curling to fill this gap . This happens only at large Mg2+ concentrations since GDP at the E-site has low affinity for Mg2+ [30] . At low Mg2+ concentrations , disassembly becomes slower and outward curling becomes less pronounced [11] . Tangential energy profiles at the longitudinal inter-dimer interface were also shown to be uneven and confirmed the hypothesis that GTP hydrolysis produces a strain which promotes sideway titling [15 , 16] . However , much of this strain could be tolerated within the lattice constraints and its orthogonality to the direction of outward curling rules out its role in disassembly . We also identified the most important residues and domains with respect to MT stability at both interfaces and their energetic contributions . At the lateral interface , the α/M-loop , β/M-loop , α/H3 helix , α/N-terminal loop and the α/H2-S3 loop were shown to be most stabilizing while the β/H3 helix was actually destabilizing . This supports predictions based on structural studies [25 , 28] . Residue α/Tyr283 was shown to form a very strong network of vdW interactions with neighboring residues and to provide the largest amount of stability at the lateral interface . At the longitudinal inter-dimer interface , the β/C-terminal domain was found to be of paramount importance not only to stability but also to the mechanism of MT disassembly . In particular , residues β/Arg401 , β/Phe404 , and β/Trp407 of the C-terminal H11 helix and the H11-H11’ loop were shown to provide more than 20% of longitudinal stability in both the GTP- and GDP-Models . The complete breakdown of MT energetics per every single residue was further analyzed in order to provide crucial insights into many aspects of MT dynamic instability . Of highest importance is the calculation of the amount of force generated through outward curling due to uneven longitudinal interactions . This could help unravel many aspect of the molecular machinery of cell division , in particular the force generation requirement for chromosome segregation . As a future prospect , simulation of a free protofilament is necessary in order to find out about the effect of uneven longitudinal energy distribution on the extent of outward curling . By comparing the energy of a free protofilament to the energy of a protofilament constrained within our MT model , we can predict the amount of free energy released by outward curling and additional light could be shed on the mechanism and driving forces in MT disassembly . Also , simulating a GDP-Taxol case is necessary to understand the molecular mechanisms by which taxol bound to an MT prevents outward curling and MT disassembly .
The recent structures for GMPCPP and GDP bound MTs at resolutions of 4 . 7 and 4 . 9 Å , respectively [15] , represented an excellent starting point for building the models presented here . The 3×3 lattice PDB structures of 3J6E ( with GMPCPP ) and 3J6F ( with GDP ) were processed using MOE software [35] by the addition of hydrogens and prediction of ionization states . The central tubulin dimer of the 3×3 lattice in each case was separated and was repaired by the addition of missing residues ( Residue 1 in β-tubulin and residues 1 , 39 to 48 , 440 in α-tubulin ) from the PDB structure 1TUB [36] , using MOE . We modified GMPCPP into GTP since in our simulations there is no need to use the nonhydrolyzable GTP analogue as hydrolysis is not expected in MD simulations . Next , for both GTP and GDP systems , the repaired tubulin was superimposed over the 13 tubulin dimers in the complete MT model built by Wells and Aksimentiev [24] , producing a hybrid complete MT model for both systems . Thus , we produced two models , the GTP-Model and the GDP-Model , by combining the helical structural configuration developed by Wells and Aksimentiev with the lattice tubulin coordinates obtained from Alushin’s model . Several clashes existed at lateral interfaces between tubulin dimers and were resolved through a short minimization using the Generalized Born ( GB ) continuum model in Amber [37] . Each model , as shown in Fig 1B , has 13 tubulin dimers in an MT orientation . For the GDP-Model , each tubulin has GTP , Mg2+ and four coordinating water molecules at the α-tubulin N-site , and GDP at the β-tubulin E-site . For the GTP model , there was GTP , Mg2+ and four coordinating water molecules at both the N-site and the E-site . Solvation was carried out using box of dimensions 293 . 85 × 293 . 85 × 83 . 38 ( or 81 . 20 ) Å3 for the GTP- and GDP-Models , respectively . The z-component was obtained from Alushin’s lattice structure [15] and ensures perfect longitudinal alignment of tubulin dimers in both systems ( see Fig 1B ) . Both x and y components were obtained from Wells’ structure [24] . A total of 181 , 000 TIP3P water molecules were added in the solvation box . This number was obtained based on several optimization trials which guaranteed consistency in box dimensions and density throughout the simulations . A total number of 442 Na+ ions was needed for neutralizing the GTP-Model , versus 455 for the GDP-Model . An extra 327 Na+ and Cl− ions were added to bring the salt concentration to 0 . 1 M . During the addition of water and ions , we made sure that no atoms were placed in positions which will be occupied by the periodic images of our system in both the positive and negative z direction ( see the gaps in the water box of Fig 1B ) . Thus , exploiting the periodic boundary conditions , the mirroring of our nearly 720 , 000-atom system in all directions should effectively result in an infinite number of infinitely long MTs , ( see S1 Movie ) . The AMBER Molecular Dynamics package was used for solvation , ionization , and dynamics [37] . The all-atom forcefield AMBERff12SB was used to parameterize the protein [38 , 39] . Cofactors were parameterized utilizing the parameter set developed by Meagher et al . [40] . Each of the two systems was then minimized through nearly 1000 steps of the steepest descent algorithm followed by about 6000 steps of the conjugate gradient algorithm . Then , the systems were heated , with restraints of 10 kcal mol−1 Å−2 on the protein , to a temperature of 310 K using the Langevin thermostat over 20 ps under constant volume . This was followed by 200 ps of density equilibration under constant temperature and pressure , in which the restraints were eliminated gradually , followed by a production phase of 50 ns for each system . Simulations were performed using NVIDIA Tesla K20X GPU cards on the PharmaMatrix Cluster ( University of Alberta ) through AMBER GPU-accelerated code [41–43] . All simulations were performed using periodic boundary conditions employing the particle-mesh Ewald method for treating long-range electrostatics and a non-bonded cut off of 10 . 0 Å under constant pressure with anisotropic pressure scaling . The 50-ns trajectory of each system was analyzed for several structural and conformational aspects . Most of the analysis was done utilizing the CPPTRAJ module in AMBER [44] , MM/GBSA implementation in AMBER [45] plus several scripts that we designed to facilitate data analysis . The software VMD 1 . 9 . 1 was also used for viewing trajectories and image rendering [46] . Data analysis included calculating the total as well as the per-residue MM/GBSA binding energies [47] between pairs of tubulin dimers in lateral and longitudinal orientations . These calculations involved all the 13 heterodimers included in the simulations and would always give the energy per MT ring ( Fig 1B ) . Hence , energetic contributions were assessed via the equation: E x = ϵ x ( R 13 L 1 ′ ) + ∑ k = 1 12 ϵ x ( R k L k + 1 ) ( 1 ) for lateral systems , and the equation: E x = ∑ k = 1 13 ϵ x ( R k L k ′ ) ( 2 ) for longitudinal systems . In both equations , Ex represents an energetic contribution of a given residue , domain or subunit x per MT ring of 13 tubulin dimers shown in Fig 1B . In Eq 1 , ϵx ( Rk Lk+1 ) is the energetic contribution of the same entity x in a subsystem composed of only tubulin k , treated as a “receptor” , and tubulin k+1 , treated as a “ligand” . ϵ x ( R 13 L 1 ′ ) does the same but at the lateral seam , taking into account the flip between α- and β-subunits . In Eq 2 , ϵ x ( R k L k ′ ) carries the same concept except that the ligand in a longitudinal subsystem is simply the periodic image of the receptor , hence the prime . Therefore , we ended up investigating 12 lateral subsystems plus 1 lateral subsystem at the seam and 13 longitudinal subsystems , for each model . An illustration of each subsystem is shown in Fig 3 . Hence , our convention in this work is that the dimer whose M-loop is involved in lateral interactions is always termed “receptor” in lateral subsystems , and the dimer whose α-tubulin is involved in longitudinal interactions is always termed “receptor” in longitudinal subsystems . This distinction was necessary since we noticed that energetic contributions can vary between tubulin dimers acting as receptors and those acting as ligands . All the energy calculations were performed on 200 evenly-spaced snapshots from the last 10 ns of the molecular dynamics trajectory where equilibration was confirmed . A solvent and solute dielectric constant of 80 and 1 , respectively , were used for electrostatics in the Amber MM/GBSA implementation . | The molecular machinery of chromosome segregation during cell division is one of the most sophisticated molecular biology mechanisms employing the interplay of different proteins and forces . The long filamentous tube-shaped microtubule structure is a central player in chromosome segregation and cell division , making it an important physiological and therapeutic target . However , the driving force for microtubule disassembly and dynamic instability , and hence force generation , is still not fully understood . In our all-atom molecular dynamics simulations we calculated the energy of interactions , within a microtubule cylinder , that is responsible for microtubule stability . We broke this energy down to individual contributions of every residue and domain . Different energy profiles enabled us to unravel the driving force behind microtubule disassembly and force generation , a longstanding unanswered biological question . We also elucidated the mechanism of disassembly and explained the effects of different factors on disassembly rates . Our list of energetic contribution of single amino acid residues could also serve in tailor-designing engineered microtubules that could be used for therapeutic and diagnostic purposes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Detailed Per-residue Energetic Analysis Explains the Driving Force for Microtubule Disassembly |
Human gammaherpesviruses are associated with malignancies in HIV infected individuals; in macaques used in non-human primate models of HIV infection , gammaherpesvirus infections also occur . Limited data on prevalence and tumorigenicity of macaque gammaherpesviruses , mostly cross-sectional analyses of small series , are available . We comprehensively examine all three-rhesus macaque gammaherpesviruses -Rhesus rhadinovirus ( RRV ) , Rhesus Lymphocryptovirus ( RLCV ) and Retroperitoneal Fibromatosis Herpesvirus ( RFHV ) in macaques experimentally infected with Simian Immunodeficiency Virus or Simian Human Immunodeficiency Virus ( SIV/SHIV ) in studies spanning 15 years at the AIDS and Cancer Virus Program of the Frederick National Laboratory for Cancer Research . We evaluated 18 animals with malignancies ( 16 lymphomas , one fibrosarcoma and one carcinoma ) and 32 controls . We developed real time quantitative PCR assays for each gammaherpesvirus DNA viral load ( VL ) in malignant and non-tumor tissues; we also characterized the tumors using immunohistochemistry and in situ hybridization . Furthermore , we retrospectively quantified gammaherpesvirus DNA VL and SIV/SHIV RNA VL in longitudinally-collected PBMCs and plasma , respectively . One or more gammaherpesviruses were detected in 17 tumors; generally , one was predominant , and the relevant DNA VL in the tumor was very high compared to surrounding tissues . RLCV was predominant in tumors resembling diffuse large B cell lymphomas; in a Burkitt-like lymphoma , RRV was predominant; and in the fibrosarcoma , RFHV was predominant . Median RRV and RLCV PBMC DNA VL were significantly higher in cases than controls; SIV/SHIV VL and RLCV VL were independently associated with cancer . Local regressions showed that longitudinal VL patterns in cases and controls , from SIV infection to necropsy , differed for each gammaherpesvirus: while RFHV VL increased only slightly in all animals , RLCV and RRV VL increased significantly and continued to increase steeply in cases; in controls , VL flattened . In conclusion , the data suggest that gammaherpesviruses may play a significant role in tumorogenesis in macaques infected with immunodeficiency viruses .
The lymphotropic human gammaherpesviruses Epstein Barr Virus ( EBV ) and Kaposi’s Sarcoma-Associated Herpesvirus ( KSHV ) are associated with malignancies in the setting of HIV infection[1] . EBV is associated with aggressive B cell Non-Hodgkin’s lymphomas such as diffuse large B cell lymphoma ( DLBCL ) and Burkitt’s lymphoma ( BL ) , as well as classical Hodgkin’s lymphoma and other lymphoproliferative disorders . KSHV is associated with Kaposi’s sarcoma ( KS ) , primary effusion lymphoma ( PEL ) and multicentric Castleman’s disease ( MCD ) . While the incidence of AIDS-defining malignancies has declined since the introduction of potent anti-retroviral therapies , they still remain an important cause of morbidity and mortality in HIV infected persons [2 , 3] and the risk of developing non-AIDS defining lymphomas and other malignancies is 2–3 fold higher in HIV infected individuals than in the general population [4 , 5] . Rhesus macaques ( Macaca mulatta ) are the monkey species most widely-used in non-human primate ( NHP ) models of HIV infection; AIDS research conducted in this NHP species has provided many important insights into HIV pathogenesis and in vivo proofs of concept in the evaluation of preventive and therapeutic approaches ( reviewed in: [6–8] ) . Three gammaherpesviruses are known to infect rhesus macaques: rhesus lymphocryptovirus ( RLCV ) , closely related to EBV; and two rhadinoviruses , retroperitoneal fibromatosis herpesvirus ( RFHV ) , closely related to KSHV and rhesus rhadinovirus ( RRV ) . Experimental infection of macaques with RLCV and RRV have been used as models for AIDS-associated malignancies caused by gammaherpesviruses [9–13] . In addition , a related virus isolated from pig tailed macaques with lymphoma was shown to cause lymphoma in rabbits[14 , 15] . However , the role of naturally occurring gammaherpesvirus infections in macaques subsequently infected experimentally with SIV/SHIV and their role in malignancies has not been extensively studied . Initial investigations at the Tulane National Primate Research Center reported on the prevalence of RLCV and RRV , but not RFHV , in healthy and SIV-infected macaques and suggested a role for RLCV , but not RRV , in SIV-associated lymphomas [16 , 17] . A prevalence study of RRV and RFHV , but not RLCV , in healthy rhesus macaques in the California National Primate Research Center breeding colony has also been reported [18] . More recently , a retrospective study of RLCV , RRV and RFHV detection by PCR in lymphoma tumor samples from the Washington National Primate Research Center suggested a role for RLCV in SIV/SHIV related B cell lymphomas and RRV in SRV2-related T cell lymphomas , but RFHV was not detected . The study reported cross sectional PCR detection in tumor tissues only and did not examine non-tumor tissues or PBMCs [19] . We sought to further elucidate the potential role of naturally occurring gammaherpesvirus infections in malignancies arising in rhesus macaques experimentally infected with SIV/SHIV , by testing for gammaherpesvirus DNA samples from animals employed in SIV and SHIV studies conducted by the AIDS and Cancer Virus Program between 2001 and 2015 , and comparing results from animals that either were or were not diagnosed with malignancies . First , we developed sensitive and specific real time quantitative PCR assays for RRV , RFHV and RLCV and used these to quantify DNA viral load ( VL ) in tumor tissue and adjacent unaffected tissues collected at necropsy from rhesus macaques that developed histologically confirmed malignancies . We then used immunohistochemistry and in situ hybridization to further characterize the malignancies occurring in these animals and their association with specific gammaherpesviruses . Finally , we determined the seroprevalence of all three gammaherpesviruses in the study animals and quantified gammaherpesvirus DNA VL in longitudinally collected PBMCs of the animals that had developed malignancies as well as SIV/SHIV infected control animals .
Characteristics of the 18 cases and 32 control study animals are shown in Table 1 . Controls were selected amongst animals euthanized for end-stage SIV/SHIV disease , whilst animals with tumors presented generally with symptomatic lesions that rendered euthanasia necessary , and that were histopathologically demonstrated to be malignant . Overall , cases and controls did not significantly differ at euthanasia in age , SIV/SHIV viral load or CD4 counts , and they were followed over a similar follow up period , although the cases were sampled more frequently; among the cases there were more females , which were significantly older ( median: 5 , interquartile intervals [IQR] 5–8 for male cases , median 16 , IQR 11–16 for female cases ) . Summary results of serological analyses are shown in Table 1 . The prevalence of antibodies against each of the three viruses was similar in cases and controls . Antibodies against RRV and RLCV were detected in 94% and 81% of cases , respectively and in 90% of controls . Prevalence of antibodies against RFHV was lower , 28% in cases and 25% in controls . The prevalence of double and triple infection is shown in S2 Fig . Formalin-fixed , paraffin-embedded ( FFPE ) blocks were available from eight macaques , six with lymphoma , one with fibrosarcoma and one with adenocarcinoma . Snap-frozen tumor specimen and frozen lymphocyte pellets from biopsies of lymphoid organs were available from an additional six and four macaques with lymphoma , respectively , for which FFPE had not been prepared . Multiple samples from each tumor and from surrounding unaffected tissues were examined . FFPE samples of four monkeys were microdissected by laser-capture or manually to obtain affected and adjacent unaffected tissues; in 11 other cases , adjacent unaffected tissue was obtained by separate sampling and identified histologically . For one animal with lymphoma no tissue specimen was available . Viral load data are summarized in Table 2 and in S1 Fig . Gammaherpesvirus DNA was detected by quantitative PCR in tumor tissues of four ( RFHV ) , 14 ( RRV ) and 16 ( RLCV ) macaques . In 11 tumors , DNA from two of the viruses was detected while in three animals , tumors DNA from all three viruses was detected . Most lymphomas appeared to be RLCV-related based on the observed high RLCV viral loads and low or absent RRV and RFHV VL in tumor tissues . One animal , with a Burkitt-like lymphoma ( BL ) had extremely high RRV load , low RLCV and undetectable RFHV in the tumor tissue , which suggests a potential RRV etiology . Similarly , the animal with fibrosarcoma had a high RFHV viral load , very low RRV and undetectable RLCV in the tumor indicating a possible role for RFHV in this case . This is consistent with the known association between RFHV and retroperitoneal fibromatosis , a multifocal fibroproliferative syndrome that arises in the peritoneum , ileocecal junction and adjacent mesenteric lymph nodes in macaques with SIV/SHIV or SRV-2 infection[20] . The only animal with a carcinoma had low or undetectable viral loads for all three viruses; it is unlikely that gammaherpesviruses contributed to tumorigenesis in this case . Gammaherpesvirus DNA was also detected in adjacent unaffected tissues , especially in lymph nodes , but levels were generally much higher in malignant tissues . In DLBCL cases , the median RLCV VL and interquartile range ( IQR ) were 4 . 0 x 102 ( 12–1 . 3 x 103 ) in tumors and 32 ( 0–1 . 3 x 105 ) in unaffected tissues , p = 0 . 02; in the fibrosarcoma case , median RFHV VL was 7 . 2 x 105 ( IQR: 6 . 7 x 105−8 . 5 x 105 ) in tumor tissue and 1 ( IQR: 0–1 , 1 x 103 ) in unaffected tissues , p = 0 . 008; in the BL-like case , median RRV VL was 6 . 8 x 106 ( IQR: 1 . 4 x 106−9 . 2 x 106 ) in the tumor and 3 . 4 x 104 ( IQR: 2 . 4 x 104−9 . 4 x 104 ) in unaffected tissues , p = 0 . 003 . Multiple samples from each tumor and from surrounding unaffected tissues were examined . We performed detailed immunohistochemistry using a panel of antibodies specific for Bcl-2 , Bcl-6 , CD3 , CD20 , c-Myc , Ki-67 and Pax5 . Histologically , 14 lymphomas were classified as diffuse large B-cell lymphoma ( DLBCL ) . One tumor was categorized as BL—like and one tumor had features of both DLBCL and BL , and was characterized pathologically as a B cell lymphoma , unclassified ( BCL-U ) [21 , 22] . Of the three remaining malignancies , one was a fibrosarcoma , one a carcinoma and one , histopathologically classified as lymphoma at necropsy , had no tissue available for further analyses . S4 Table shows the sites , gross exam , histologic and in situ hybridization findings made in all the tumors tested . A representative DLBCL case , and the single BL-like and fibrosarcoma cases are shown in Figs 1–3 . Lymphoma tissues consisted of different populations of lymphocytes , centroblasts , and immunoblasts , many with plasmacytoid differentiation with heterogenous expression of both CD20 and Pax5 , showing tumor cells at different stages of differentiation . This was also evident from the variable presence of euchromatin , prominent nucleoli , and frequency of mitotic figures . All DLBCLs were strongly positive for Ki-67 , c-Myc , and Bcl-2 but negative for Bcl-6 ( Fig 1 ) while the BL-like case was positive for Bcl-6 but low for Bcl-2 [23] ( Fig 2 ) . The tumor diagnosed as colonic fibrosarcoma was composed of elongated spindle cells with infiltration of many neutrophils ( Fig 3 ) . The neoplastic cells were negative for desmin but expressed abundant vimentin and collagen I . ( Fig 3 ) . To further explore a potential direct role of gammaherpesvirus infection in these tumors , we performed immunohistochemistry on tissues using antibodies specific for the EBV LMP1 and EBNA1 , that have been demonstrated to be cross-reactive with the RLCV orthologs , as well as anti-RRV major capsid protein ( clone 3D1 . 2 ) , and anti-KSHV ORF73 , with known cross-reactivity to the RFHV ortholog [24] . In addition , we performed RNAscope in situ hybridization using probes designed to specifically target RNA from RLCV , RRV ( Figs 1–3 ) and RFHV . High RLCV RNA expression was seen in all DLBCLs but not in the BL-like case ( Figs 1 and 2 ) . Interestingly , cells with low levels of cytoplasmic RLCV LMP1 were seen in both non-malignant and B-cell lymphoma ( BCL ) tissues , however , few cells with RLCV EBNA1 were found in the non-malignant tissues . In addition , only rare RRV positive cells were seen in DLBCL tissues ( Fig 1 ) . In stark contrast to DLBCL cases , elevated levels of RRV and no RLCV were seen in the one BL-like case ( Fig 2 ) . In the fibrosarcoma , only tumor cells were positive for KSHV ORF73 , while all other BCL and non-malignant tissues were negative or low ( Fig 3 ) . In the fibrosarcoma tumor case , we were not able to detect RRV RNA or protein ( Fig 3 ) . Collectively , these findings are in line with our tissue PCR results and strongly suggest association of RLCV with DLBCLs , RRV with a BL-like tumor , and RFHV with a KS-like tumor in NHPs . Median RRV and RLCV DNA viral load were statistically significantly higher in PBMCs of cases than controls . RFHV was detectable in PBMCs of few animals and viral load did not differ between cases and controls ( Table 1 ) . In a multivariate longitudinal analysis including age at necropsy , sex , SIV/SHIV plasma VL , CD4 counts , and gammaherpesvirus PBMC viral loads , SIV/SHIV plasma VL and RLCV DNA viral load in PBMCs were independently associated with odds of developing cancer but RRV and RFHV DNA viral loads in PBMCs were not . ( Table 3 ) . Univariate non-parametric local regressions showed that the longitudinal pattern of VL levels in the PBMCs of cases and controls differed for each virus ( Fig 4 ) . For RFHV , PBMC DNA viral load increased only slightly in both cases and controls from SIV infection to necropsy . For RLCV and RRV DNA , viral loads increased significantly in both groups upon SIV infection and continued to increase steeply in cases until diagnosis and necropsy , whereas in controls , the VL flattened . This pattern is consistent with the detection of high levels of RLCV in most of the malignant tissues in the study while the pattern seen for RRV VL is more unexpected since only one animal had an RRV VL in malignant tissue .
Infections with lymphotropic gammaherpesviruses are prevalent in rhesus macaques used in HIV research and yet are rarely evaluated , except in limited specific pathogen free colonies stringently bred , tested , and cared for to exclude these agents . In our retrospective study , antibodies to RLCV and RRV were present in >90% of cases and controls , while antibodies to RFHV were detected in 25% of the cases and 28% of the controls . These serological data are similar to previous reports from US primate centers [16–18] . Our objectives for this study , which was more extensive than most prior published surveys , and included longitudinal quantitative PCR analysis for all three rhesus herpesviruses as well as histologic ( IHC/ISH ) analysis comparing tumor tissue and corresponding non-tumor tissue , were to elucidate the potential role of naturally acquired gammaherpesviruses in malignancies occurring in SIV/SHIV infected macaques . We have found that most such malignancies were associated with RLCV , consistent with previous reports . [16 , 17 , 19] . We did observe however , a BL-like tumor with a very high RRV viral load , low or absent RFHV and RLCV , and IHC staining consistent with a RRV etiology . We also observed a fibrosarcoma in which immunohistochemistry and viral load in the tumor were consistent with an etiological role for RFHV . Thus , we show that all three gammaherpesviruses may have oncogenic potential in the setting of experimental SIV/SHIV infection . The histological classification of these malignancies observed in SIV/SHIV-infected macaques broadly resembles that of malignancies seen in HIV-infected persons . Retrospective longitudinal analyses of gammaherpesvirus load in PBMC samples from animals that did develop lymphomas showed a marked increase of RLCV DNA viral load in cases from SIV/SHIV infection to diagnosis , whereas in controls it eventually decreased , consistent with the association between RLCV reactivation and lymphoma . All the malignancies , except the carcinoma , were diagnosed at or shortly before necropsy , therefore it is unclear from our data whether elevated levels of RLCV preceded the development of lymphoma or accompanied it . Prospective analyses conducted in future NHP studies will be needed to further elucidate this dynamic . Interestingly , whilst not an independent risk factor , RRV viral load also increased over time from SIV/SHIV infection in all animals developing lymphoma , whereas it eventually decreased in controls , even though , among the more limited subset of animals seropositive for this gammaherpesvirus , only one animal developed a lymphoma showing exclusively RRV in tumor tissue . On the contrary , in infected animals , RFHV DNA VL did not show differential kinetics between cases and controls . Further studies in additional animals will be necessary to understand the relationship between RRV replication and the pathogenesis of lymphomas associated with RLCV . Besides RLCV DNA VL , high SIV/SHIV RNA VL was independently associated with cancer risk , emphasizing the role of progressive SIV/SHIV infection and accompanying immunodeficiency . While provocative , the retrospective , correlative nature of these data does not permit the direct demonstration of a causative etiologic role for these viruses in the observed malignancies . Further , the retrospective nature of the results and limited samples available from studies that were conducted with other primary objectives do not allow us to directly address the role of progressive SIV/SHIV virus associated immunodeficiency and , potentially , declines in specific immune responses to the gammaherpesviruses in contributing to development of tumors . Similarly , the potential role of an immunoinflammatory milieu engendered by uncontrolled AIDS virus replication and associated host responses in breaking gammaherpesvirus latency remains unaddressed by the present results . However , our observations extend previous work in important ways by providing longitudinal and comparative data for all three rhesus gammaherpeseviruses in the same animals , along with detailed comparative analysis using serological assays and quantitative PCR , as well as immunohistochemical and in situ hybridization analyses of both tumor and control tissues . While falling short of a “smoking gun” , in aggregate , the results strongly implicate rhesus gammaherpesviruses in contributing to the development of tumors in the setting of experimental SIV/SHIV infection and point to the importance of careful further studies to address questions of etiologic roles and mechanisms of pathogenesis . Our study suggests that RLCV and RRV are likely to play a significant role in lymphomagenesis in SIV/SHIV infected macaques and that the contribution of gammaherpesviruses to SIV-associated malignant disease is worthy of further study , particularly as these tumors recapitulate many important features of malignancies that continue to arise in HIV-infected humans , malignancies for which an animal model to evaluate novel treatment approaches would be valuable .
Specimens from eighteen animals with malignancies ( 16 with lymphoma , one with carcinoma and one with fibrosarcoma ) were available from studies conducted with a variety of different primary objectives . All selected animals were rhesus macaques ( Macaca mulatta ) of Indian origin and were infected with SIV or chimeric simian-human immunodeficiency virus ( SHIV ) . Malignancies were demonstrated at necropsy in all cases . Thirty-two control animals without malignancies were also selected from the same studies; for these animals , the indication for euthanasia was end-stage SIV/SHIV disease . Plasma and peripheral blood mononuclear cells ( PBMCs ) were prepared from whole blood collected in EDTA anticoagulated Vacutainer tubes ( BD ) . Plasma was separated from whole blood by centrifugation , recentrifuged to eliminate cells , platelets and debris , then aliquoted and then stored at -80°C until analysis . PBMCs were isolated from whole blood by Ficoll-Paque Plus ( GE Healthcare ) gradient centrifugation . Portions of isolated PBMCs were pelleted in a microcentrifuge and all liquid was removed prior to storage at -80°C . Plasma was tested for gammaherpesvirus serology at necropsy as described below . Stored PBMCs collected up to 4186 days prior to necropsy were available from a median of 11 time points ( interquartile range [IQR}8–13 ) for cases and 6 time points for controls ( IQR 5–12 ) . Formalin fixed , paraffin embedded ( FFPE ) blocks or snap frozen/cryopreserved specimen of tumors and surrounding unaffected tissues were available for 17 of the cases . Antibodies to RRV were detected using peptide based ELISAs ( ORF 65 , R8 . 1 and ORF 73 ) as previously described [24] . Antibodies to RLCV were detected using peptide based ELISAs ( VCA ) as previously described [25] , [26] . Antibodies to RFHV were detected based on ELISA cross reactivity to the recombinant ortholog KSHV proteins K8 . 1 and ORF 73 , as previously described[27] . DNA from frozen tissue samples was extracted using Trizol according to the manufacturer’s instructions after tissue homogenization using a gentleMACS dissociator ( Miltenyi Biotec ) . DNA from FFPE tissues was extracted using the phenol-based AutoGenprep 245T Animal Tissue DNA Extraction Kit ( Autogen ) according to the manufacturer’s method . DNA was extracted from PBMC pellets using QIAamp DNA blood mini kit ( QIAGEN ) according to the manufacturer’s instructions . Yield and purity were determined by NanoDrop 1000 spectrophotometer ( NanoDrop Technologies ) . DNA was stored at -20°C until subsequent assays/analyses . To develop plasmids for assay standard curves , animals seropositive for each virus were identified and nested PCR was used to amplify DNA from corresponding PBMC samples with primers specific for each virus ( Jumpstart ReadyMix , Sigma ) using cloning primers listed in S1 Table . Cloning primers were designed using GeneRunner software ( version 4 . 0 . 9 . 56 Beta ) and synthesized by Eurofins-Operon . PCR products of the expected size were excised from gels and purified ( QIAquick , Qiagen ) . Products were cloned into Promega’s T-Easy vector system II ( Promega ) and sequenced using a 3130XL Applied Biosystems sequence detection system ( Thermo Scientific ) . For each virus , sequences were obtained from multiple seropositive animals and aligned with available reference sequences from GenBank . Consensus sequences were generated and used to design primers and probes for real-time quantitative PCR ( qPCR ) using Primer Express software version 3 . 0 . 1 ( Applied Biosystems , Thermo Scientific ) . In the case of RRV , reported sequence variation of the glycoprotein B region was incorporated into the design [28] . Plasmids were purified and concentrated in a separate building by the Protein Expression Laboratory ( Leidos Biomedical , Frederick National Laboratory for Cancer Research ) to prevent template contamination of ACVP laboratory areas . All probes were labeled with FAM ( reporter ) and TAMRA ( quencher ) . Primers and probes are listed in S1 Table . Plasmid stocks were quantitated by Nanodrop 1000 ( Thermo Scientific ) and 10-fold serial dilutions were made using 1 X TE buffer pH 7 . 0 ( Ambion ) with 0 . 1 μg/ml fish sperm DNA ( Ambion ) resulting in standard curve linear dynamic ranges from 10^6 to 10^0 copies . Three standard curve dilution series were made for each assay and tested in triplicate reactions across 10 individual plates to determine inter-assay variability as shown in S2 Table . Each assay was optimized using universal master mix ( Applied Biosystems ) with final primer/probe concentrations of 100 nM . The cycling conditions for all assays consisted of a 2-minute hold at 50°C , 95°C hold for 10 minutes followed by 45 cycles performed at 95°C for 15 seconds , 55°C for 30 seconds , and annealing at 60°C for 1 minute . SIV/SHIV RNA viral loads in plasma were determined by quantitative qRT-PCR as previously described[29] , with progressively improved threshold sensitivity of the assay over the period covered by the study through refinements of the assay as described[30] . Quantitative PCR for rhesus CCR5 was performed largely as previously described [31] in order to determine the number of cell-equivalents . The CCR5 probe was adapted with FAM/TAMRA modifications and ran as a single-plex qPCR assay on an Applied Biosystems 7900 HT sequence detection system . Inter-assay standard curve performance in shown in supplemental Table 2 . Serial 7 μm sections were cut and individually placed onto positively charged glass microscope slides . Tumor and unaffected tissue regions , as annotated by a pathologist , were dissected without the removal of the paraffin . Using a single edge razor blade and a dissecting microscope , unwanted tissue was removed from slides and discarded . The unaffected tissue from all slides of one specimen were removed and placed into a single 1 . 5 ml microcentrifuge tube for DNA extraction . Then the tumor tissue from the same specimen was removed from all slides and placed into another 1 . 5 ml microcentrifuge tube . Laser capture microdissection ( LCM ) and collection of annotated targets from serial 7 μm FFPE sections was performed on a MMI CellCut Plus microdissection instrument ( MMI Molecular Machines&Industries , Glattbrugg , Switzerland ) with the following settings: laser speed-37% , laser focus-78% and laser power-41% . LCM workflow , LCM slide preparation , target dissection and collection were carried as previously described for FFPE NanoString LCM sample[32] DNA extraction was performed using the DNA extraction protocol used for hand-micro dissected samples . Cell line controls for IHC were developed using B cell lines: Ramos ( virus negative ) ; BCBL-1 ( KSHV positive ) ; Namalwa ( EBV positive ) ; LCL 8664 ( RLCV positive ) ; and BJAB-RRV ( RRV positive ) . BJAB-RRV was a kind gift from Dr . Blossom Damania , UNC Chapel Hill . Approximately 10 million cells from each cell line were pelleted and the cell pellets were fixed in 4% PFA overnight at room temperature then wash in 80% ethanol , mixed with HistoGel ( HG-400-012 , Thermo Scientific ) and stored at 4°C until processing for paraffin embedding . Infected and uninfected cell lines were embedded into one block for direct comparison . FFPE sections were dewaxed in xylenes and rehydrated with serial washes of ethanol to water . Heat inducted epitope retrieval was performed in 0 . 01% citraconic anhydride buffer ( pH 7 . 4 ) in a Decloaker pressure cooker ( Biocare Medical , Inc . ) programed for 30 seconds at 122°C . After slides were cooled to room temperature , rinsed for 5–10 min in ddH2O , slides were incubated for 30min in blocking buffer [1x TBS-Tween20 ( 0 . 05% ) containing 0 . 25% casein protein] to block non-specific staining . After removing the blocking solution , the slides were incubated with primary antibodies diluted in blocking solution overnight at 4°C or RT . Slides were placed in wash buffer ( 1x TBS-Tween20 ( 0 . 05% ) ) for 5 min and endogenous hydrogen peroxidases were quenched by incubating slides in 1 . 5% hydrogen peroxide in TBS buffer for 10 min . Detection of gammaherpesvirus specific antibodies was performed using the mouse or rabbit Polink polymer staining system ( Golden Bridge International , Inc ) according to manufacturer’s instructions then Impact DAB ( 3 , 3′-diaminobenzidine; Vector Laboratories ) as previously reported [33] . Slides were counterstained with hematoxylin and mounted with Permount . Details of primary antibodies and used dilutions are shown in S3 Table . Double staining for CD3 and CD20 was done with Warp Red ( Biocare Medical , Inc . ) as chromogen for CD3 ( red ) and DAB for CD20 ( brown ) . Slides were scanned at high magnification ( × 200 ) using a whole-slide scanning microscope ( Aperio AT2 System , Aperio Technologies ) , yielding high-resolution data from the entire tissue section [33] . RNAscope in situ hybridization was performed as described previously [34]Briefly , following HIER ( Pretreat 2 step; Advanced Cell Diagnostics , ACD ) and proteinase digestion ( Pretreat 3 step , ACD ) , the slides were incubated 2 hours at 40°C with either rLCV ( ACD-ref:448011 ) , RRV ( ACD-ref:448021 ) or RFHV ( ACD-ref:448031 ) probe . Amplification steps were performed according to the ACD protocol with the exception that all wash steps used a 0 . 5X wash buffer . Slides were scanned at high magnification ( ×400 ) using a whole-slide scanning microscope ( Aperio AT2 System , Aperio Technologies ) , yielding high-resolution data from the entire tissue section . A Mann-Whitney test was used to compare mean log viral load and other continuous variables between cases and controls in univariate analyses , and equality of proportions was tested using large-sample statistics . Hierarchical linear models were used for comparing viral load in tumor and unaffected tissues across multiple animals . Univariate non-parametric regressions ( LOWESS ) and multivariate longitudinal random effects models were used for longitudinal analyses . All statistical analyses were performed using Stata v13 . This study made use of samples from Indian-origin rhesus macaques that were housed at the National Institutes of Health ( NIH ) and cared for in accordance with the Association for the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) standards in an AAALAC-accredited facility , and all procedures were performed according to protocols approved by the Institutional Animal Care and Use Committee of the National Cancer Institute ( Assurance #A4149-01 ) and adhered to the standards of the NIH “Guide for the Care and Use of Laboratory Animals” ( National Research Council . 2011 . Guide for the care and use of laboratory animals , 8th ed . National Academies Press , Washington , DC ) . Twenty-six purpose-bred Indian-origin rhesus macaques ( Macaca mulatta ) weighing on average 7kg ( range 5-9kg ) were housed at the National Institutes of Health ( NIH ) and cared for in accordance with the Association for the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) standards in an AAALAC-accredited facility and all procedures were performed according to protocols approved by the Institutional Animal Care and Use Committee of the National Cancer Institute ( Assurance #A4149-01 ) . Animals were maintained in Animal Biosafety Level 2 housing with a 12:12-hour light: dark cycle , relative humidity 30% to 70% , temperature of 23 to 26°C and all animals were observed twice daily by the veterinary staff . Filtered drinking water was available ad libitum , and a standard commercially formulated nonhuman primate diet was provided thrice daily and supplemented 3–5 times weekly with fresh fruit and/or forage material as part of the environmental enrichment program . Environmental enrichment: Each cage contained a perch , two portable enrichment toys , one hanging toy , and a rotation of additional items ( including stainless steel rattles , mirrors , and challenger balls ) . Additionally , the animals were able to listen to radios during the light phase of their day and were provided with the opportunity to watch full-length movies at least three times weekly . At the start of the study , all animals were free of cercopithecine herpesvirus 1 , simian immunodeficiency virus ( SIV ) , simian type-D retrovirus , and simian T-lymphotropic virus type 1 . All animals were treated with enrofloxacin ( 10 mg/kg once daily for 10 days ) , paromomycin ( 25 mg/kg twice daily for 10 days ) , and fenbendazole ( 50 mg/kg once daily for 5 days ) followed by weekly fecal culture and parasite exams for 3 weeks to ensure they were free of common enteric pathogens . At least a 4-week post treatment period allowed time for stabilization of the microbiome prior to use in this study . Physical Examination: All animals received complete physical examinations during preventative healthcare . Examinations were performed under anesthesia , generally using ketamine ( 10–25 mg/kg , IM ) , telazol ( 4–10 mg/kg , IM ) , and/or dexmedetomidine ( 7 . 5–15 μg/kg , IM ) , and generally concurrently with other procedures ( e . g . phlebotomy ) to reduce the total number of anesthesia events required . During all procedures , animals were monitored by vet technical staff . All monkeys were observed at least twice daily by trained veterinary technical staff for any abnormal signs or behaviors . Supportive treatment was administered as suggested by the clinical veterinarian . Animal Procedures: All procedures were performed using chemical restraint unless specifically mentioned otherwise to ensure the safety of both staff and animals . Choice of anesthetic depended on the procedure , but was generally performed using ketamine ( 10–25 mg/kg , IM ) , telazol ( 4–10 mg/kg , IM ) , and/or dexmedetomidine ( 7 . 5 15 g/kg , IM ) . All of these drugs are commonly used in nonhuman primates and are considered safe and effective . For euthanasia , animals were initially sedated with ketamine ( 10–25 mg/kg , IM ) or telazol ( 4–10 mg/kg , IM ) followed by an overdose of sodium pentobarbital ( >75 mg/kg , IV ) to effect , in accordance with ACUC guidelines . | Human gammaherpesviruses cause malignancies in HIV infected persons; in SIV infected macaques , gammaherpesvirus infections also occur . To understand the potential role of the rhesus gammaherpesviruses , RRV , RFHV and RLCV in cancers occurring in monkeys during SIV and SHIV studies spanning the last 15 years , we developed assays to measure the DNA viral load ( VL ) of each virus in these tumors and unaffected macaque tissues . We further characterized the tumors using immunohistochemistry and in situ hybridization , and quantified gammaherpesvirus DNA VL in PBMCs collected longitudinally during the original studies . We examined 18 animals with tumors and 32 controls . In all tumors , we detected one or more gammaherpesviruses; generally , one virus was predominant and very abundant compared to surrounding non-tumor tissues . SIV RNA VL in plasma and RLCV VL in PBMCs were independently associated with cancer risk . The longitudinal patterns of gammaherpesviruses VL , from SIV infection to death differed in cases and controls: while RFHV VL increased only slightly in all animals , RLCV VL and RRV VL increased significantly and continued to increase in cases , but flattened in controls . These data suggest that gammaherpesviruses may play a significant role in tumorogenesis in macaques infected with immunodeficiency viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"microbiology",
"vertebrates",
"cancers",
"and",
"neoplasms",
"animals",
"mammals",
"retroviruses",
"viruses",
"primates",
"oncology",
"hematologic",
"cancers",
"and",
"related",
"disorders",
"immunodeficiency",
"viruses",
"animal",
"models",
"rna",
"viruses",
"dna",
"viruses",
"experimental",
"organism",
"systems",
"lymphomas",
"herpesviruses",
"viral",
"load",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"rhesus",
"monkeys",
"infectious",
"diseases",
"fibrosarcoma",
"monkeys",
"medical",
"microbiology",
"microbial",
"pathogens",
"sarcomas",
"hematology",
"siv",
"macaque",
"eukaryota",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"lentivirus",
"amniotes",
"organisms"
] | 2018 | Gammaherpesvirus infection and malignant disease in rhesus macaques experimentally infected with SIV or SHIV |
Plant pathogenic fungi cause massive yield losses and affect both quality and safety of food and feed produced from infected plants . The main objective of plant pathogenic fungi is to get access to the organic carbon sources of their carbon-autotrophic hosts . However , the chemical nature of the carbon source ( s ) and the mode of uptake are largely unknown . Here , we present a novel , plasma membrane-localized sucrose transporter ( Srt1 ) from the corn smut fungus Ustilago maydis and its characterization as a fungal virulence factor . Srt1 has an unusually high substrate affinity , is absolutely sucrose specific , and allows the direct utilization of sucrose at the plant/fungal interface without extracellular hydrolysis and , thus , without the production of extracellular monosaccharides known to elicit plant immune responses . srt1 is expressed exclusively during infection , and its deletion strongly reduces fungal virulence . This emphasizes the central role of this protein both for efficient carbon supply and for avoidance of apoplastic signals potentially recognized by the host .
Plant pathogenic fungi cause major yield losses and affect the quality and safety of food and feed produced from infected plant material . Different fungi have developed different strategies to deal with their hosts . Infected plants are either kept alive to ensure a prolonged supply of organic carbon and other compounds to the pathogen ( biotrophic fungi ) , or they are destroyed and the pathogen feeds on dead or dying plant tissue ( necrotrophic fungi ) . Other fungi start with a biotrophic infection and switch to necrotrophic behavior at later stages of infection or under certain environmental conditions ( hemibiotrophic fungi ) . Recognition of such pathogens by infected plants typically results in the production of reactive oxygen species and in hypersensitive cell death [1] . Obviously , plant defense responses resulting in hypersensitive cell death will be very effective against biotrophic fungi , whereas necrotrophic pathogens might even benefit from host cell death , and in fact , plants use different defense responses for biotrophic and necrotrophic fungi [1] , [2] . The most important challenge for all pathogens is , therefore , the development of strategies allowing the avoidance of signals potentially recognized by the host . The basidiomycete U . maydis is a ubiquitous pathogen of maize ( Zea mays ) , one of the world's most important cereal crops [3] . As a biotrophic fungus , U . maydis depends on living plant tissue and does not use aggressive virulence strategies [4] . During the infection process , fungal hyphae traverse plant cells without eliciting apparent host defense responses , a prerequisite for successful infection and persistent growth and development of a biotroph on its live host . U . maydis hyphae invaginate the plasma membranes of invaded plant cells , resulting in narrow contact zones that are perfectly suited for the uptake of organic carbon by the fungus [5] . Infections with U . maydis lead to the formation of tumors that consist of proliferating plant cells and of fungal hyphae ( Figure 1A and 1B ) . Comparisons of transcript and metabolite levels in U . maydis-infected with noninfected maize leaves revealed an inhibition or delay in the sink-to-source transition of infected leaves [6] , [7] , which is in line with the increased carbon demand of the forming tumor . All transport proteins identified so far in symbiotic or pathogenic fungus/plant interactions are specific for monosaccharides [8]–[10] and catalyze the uptake of glucose or fructose and , to a lesser extent , of other hexoses . It was speculated that these hexose transporters act in combination with fungal and/or plant-derived cell wall invertases [11] , [12] to supply the pathogen with carbon derived from extracellular sucrose hydrolysis . The impact of these transporters on the development of fungal pathogens within the host plant has never been proven . However , plants have evolved mechanisms to sense extracellular ( apoplastic ) changes in glucose concentrations , e . g . , produced from extracellular sucrose hydrolysis , and respond to these changes with the induction of defense responses [12]–[16] . Thus , feeding strategies avoiding invertase-derived glucose production in the apoplast might by advantageous especially for biotrophic fungi . Here , we present the identification and functional characterization of Srt1 , a novel high-affinity , sucrose-specific transporter from the biotrophic fungus U . maydis . We show that Srt1 represents a virulence factor essential for the successful development of the fungus within its host , as infections of maize with Δsrt1 strains result in strongly reduced disease symptoms . The successful infection of maize by U . maydis without induction of defense responses is likely to result from an efficient competition of the U . maydis Srt1 protein with the low-affinity plant sucrose transporters for apoplastic sucrose , and potentially from the avoidance of apoplastic glucose signaling .
Compared to the progenitor strain SG200 , a solopathogenic strain that can infect corn plants without a mating partner [17] , U . maydis strains deleted for srt1 ( SG200Δsrt1 ) did not show altered growth on agar medium supplemented with different carbon sources ( Figure 2A to 2D ) . This is in line with the observation that srt1 expression is not detected under these conditions ( Figure 3A ) . Moreover , the fact that srt1 expression is not induced on medium without any carbon source demonstrates that it is not regulated by catabolite repression . In contrast , growth of wild-type U . maydis in planta results in a rapid induction of srt1 expression ( Figure 3A ) . Expression reaches a maximum at 4 to 8 days post infection ( dpi ) when most hyphae have reached the vascular bundles to spread inside the plant and when tumor formation is initiated . During earlier stages of infection only weak expression of srt1 was observed ( Figure 3A ) . This suggests that plant-derived signals are needed for srt1 expression and that Srt1 might play a pivotal role in U . maydis/maize interaction . These results were confirmed in analyses with a modified SG200 strain ( SG200Δsrt1::srt1-GFP ) in which the native srt1 gene was replaced by an srt1-GFP fusion . Microscopic analysis of this strain revealed no fluorescence when cells were grown on minimal medium with 1% glucose ( Figure 3B ) or 1% sucrose ( Figure 3C ) . After infection of maize leaves , however , a distinct GFP signal at the cell periphery was observed ( Figure 3D ) . This ( 1 ) corroborates the plant-specific expression of srt1 and ( 2 ) suggests a plasma membrane localization of the protein . Plant infection experiments with SG200 and SG200Δsrt1 revealed major differences . Whereas infections with SG200 caused massive tumor formation ( Figure 4A and 4B ) , infections with SG200Δsrt1 resulted only in marginal disease symptoms . In most cases , infected plants showed no symptoms , only chlorotic lesions , or minute tumors ( Figure 4A and 4B ) . Moreover , strain SG200Δsrt1::srt1-GFP which had been used for the analyses shown in Figure 3B to 3D displayed similar infection rates and symptom development as the wild-type strain , demonstrating that the srt1-GFP fusion encodes a functionally active Srt1-GFP protein . To exclude the possibility that the observed loss of virulence in SG200Δsrt1 mutants ( Figure 4A and 4B ) resulted from indirect effects and not from a loss of srt1 , the srt1 deletion mutant was complemented with an srt1 wild-type copy . The resulting strain , SG200Δsrt1-srt1::ip , displayed similar infection rates and symptom development as SG200 or SG200Δsrt1::srt1-GFP . This confirmed that the observed reduced virulence of SG200Δsrt1 mutant strains results from the loss of srt1 . With respect to tissue colonization , SG200Δsrt1 hyphae did not differ from SG200 hyphae at the different developmental stages during disease progression ( Figure S2 ) . The intronless srt1 gene encodes a protein of 546 amino acids . The Srt1 protein has 12 predicted transmembrane domains ( TMDs [18] ) and a large extracellular loop between TMD1 and TMD2 ( Figure 1C ) , a typical structural feature of previously characterized fungal and plant hexose transporters [8] , [19] . Sequence comparisons revealed a moderate similarity ( less than 30% identity ) of Srt1 to a large group of transport proteins ( Figure S3 ) that includes numerous well-characterized high-affinity monosaccharide transporters from plants and fungi as well as some low-affinity maltose transporters from Saccharomyces cerevisiae [20]–[22] , Pichia angusta ( synonym: Hansenula polymorpha [23] ) , or Schizosaccharomyces pombe [24] . Phylogenetic analyses revealed that Srt1 is most closely related to a small group of so-far uncharacterized proteins ( Figure S3 ) . This group contains uncharacterized transporters from different Aspergillus species ( up to 47% identity ) and from two biotrophic relatives of U . maydis , Sporisorium reilianum ( 88% identity ) and Ustilago hordei ( 81% identity ) . To functionally characterize Srt1 , the gene was expressed in the monosaccharide transport–deficient S . cerevisiae strain EBY . VW4000 [25] , and uptake was analyzed with radiolabeled putative substrates ( d-glucose , d-fructose , d-ribose , d-xylose , d-galactose , mannitol , sorbitol , xylitol , myo-inositol ) . As Srt1 did not catalyze the uptake of any of these compounds , additional tests were performed with 14C-sucrose and 14C-maltose . Because the S . cerevisiae strain EBY . VW4000 encodes an extracellular invertase that slowly hydrolyzes extracellular sucrose , these studies of Srt1 had to be performed in the invertase-deficient S . cerevisiae strain SEY2102 [26] . In fact , transport activity could be measured with 14C-sucrose ( Figure 5A ) , but no uptake was observed for 14C-maltose ( Figure S4 ) . In competition analyses with an excess of unlabeled maltose ( an alternative substrate of plant sucrose transporters ) , trehalose ( an alternative substrate of S . cerevisiae maltose transporters ) , raffinose ( an alternative substrate of the sucrose-hydrolyzing enzyme invertase ) , or sucrose ( as positive control ) , raffinose was the only alternative compound that caused a minor inhibition of sucrose uptake ( Figure 5B ) . No transporter described so far , not even the very well-characterized sucrose transporters from higher plants [27] , showed such an extreme specificity for the disaccharide sucrose . In fungi , sucrose transport activities were so far only described as side activities of broad-specificity , low-affinity maltose or maltotriose transporters [24] , [28] . In uptake analyses in S . cerevisiae and with a wide range of different sucrose concentrations , the KM of Srt1 for sucrose was found to be 26±4 . 3 µM ( Figure 5C ) . Thus , the affinity of Srt1 for sucrose is several 100-fold to several 1 , 000-fold higher than that of the fungal maltose/maltotriose transporters [24] , [28] . Moreover , its affinity is also much higher than that of higher plant sucrose transporters ( 20-fold to more than 200-fold ) , which catalyze sucrose uptake with KM values in the millimolar range [23] . For the S . cerevisiae strain SEY2102 , d-glucose represents the primary carbon source that can be both imported and metabolized . In contrast , sucrose can be imported when srt1 is expressed , but it cannot be hydrolyzed due to a lack of invertase activity [26] . Therefore , if Srt1-mediated sucrose uptake is energy-dependent , the available energy might become limiting and the determined sucrose transport rates might be submaximal . In fact , the simultaneous presence of 14C-sucrose and glucose as metabolizable energy source strongly enhanced sucrose uptake ( Figure 6A ) , which is indicative for an energy-dependent transport . In addition to this glucose-enhanced sucrose uptake , both the clear optimum of Srt1-driven sucrose transport at acidic pH values ( Figure 6B ) as well as the sensitivity to the protonophore carbonylcyanide m-chlorophenylhydrazone ( CCCP; Figure 6C ) underline that Srt1 is an active , energy-dependent H+-symporter . These activities of plant sucrose transporters can be inhibited very specifically by the SH-group inhibitor p-chloro-mercuribenzene sulfonate ( PCMBS ) that does not affect plant hexose transporters [29] . In fact , the specificity of this inhibitor is so high that sucrose fluxes and phloem loading can be inhibited by PCMBS in whole plant or in intact plant tissues [30] . Srt1 is not inhibited by PCMBS ( Figure 6C ) . Expression of srt1 in an S . cerevisiae strain ( DBY2617 ) that possesses a cytoplasmic but no secreted invertase [31] enabled this strain not only to import 14C-sucrose , but also to grow efficiently on sucrose as sole carbon source ( Figure S5 ) . This proves that Srt1 activity alone is sufficient to meet the carbon import requirements of these cells . Thus , Srt1 is a high-affinity , high-capacity transporter that catalyzes the uptake of sufficient sucrose to fuel the growth of fungal cells . Additional analyses of the subcellular localization in S . cerevisiae with a functional Srt1::GFP fusion protein demonstrated that , as expected from the transport assays ( Figures 4 and 5 ) and complementation analysis ( Figure S5 ) , Srt1::GFP localizes exclusively to the plasma membrane ( Figure 7 ) . To validate that sucrose uptake is the primary function of Srt1 during biotrophic growth , we tested whether another transporter with a well-characterized sucrose uptake activity can functionally replace Srt1 . We selected the sucrose transporter AtSUC9 from Arabidopsis thaliana [32] . This plant transporter is plasma membrane localized , transports sucrose and maltose , and is sensitive to CCCP and PCMBS . Moreover , AtSUC9 has a KM-sucrose of 0 . 5 mM [32] , which is quite low for a plant sucrose transporter but still 20-fold higher than the KM-sucrose of Srt1 ( Figure 5C ) . In strain SG200Δsrt1::AtSUC9 , the AtSUC9 cDNA was inserted into the srt1 locus . Figure 4 demonstrates that infections with SG200Δsrt1::AtSUC9 are indistinguishable from wild-type infections with respect to tumor formation and frequency . Thus , the virulence of SG200Δsrt1 can be restored by the expression of plant sucrose transporter cDNA AtSUC9 .
The primary physiological functions of plant sucrose transporters are the loading of sucrose into the phloem or the loading of sucrose into storage vacuoles , two processes that depend on the accumulation of high sucrose concentrations ( up to 2 M ) on one side of the respective membrane [27] . Uptake beyond a certain maximum is subject to feed back inhibition and total inactivation of sucrose transport . These activities of plant sucrose transporters can be inhibited very specifically by the SH-group inhibitor PCMBS that neither affects plant hexose transporters [29] nor Srt1 ( Figure 6C ) . This is in accordance with the closer phylogenetic similarity of Srt1 to plant and fungal hexose transporters . Srt1 is a transporter that imports sucrose for immediate consumption . Accumulation of high intracellular concentrations of sucrose in U . maydis is unlikely to occur . In invertase-deficient srt1-expressing S . cerevisiae cells , imported sucrose is not hydrolyzed , and Srt1 can , therefore , accumulate sucrose to concentrations higher than in the extracellular medium ( more than 60-fold higher in Figure 6D ) . In contrast to plant sucrose transporters , the plateau of Srt1-mediated sucrose accumulation does not result from feed back ( “shut-off” ) inhibition of sucrose uptake , but rather from an equilibrium of sucrose influx and sucrose efflux , a typical property of transporters that do not accumulate their substrates under physiological conditions [33] , [34] . In summary , Srt1 appears to be the prototype of a novel sucrose transporter that is unique with regards to its high specificity and its high affinity for sucrose , and that differs significantly in its functional behavior from sucrose transporters of higher plants . The primary long-distance transport and storage form of assimilated carbon in most higher plants , including maize , is sucrose . Apoplastic sucrose concentrations were determined in several dicot plants and are typically in the low-millimolar range [35] . Thus , a transporter with the properties of Srt1 represents a perfect tool for a biotrophic fungus that resides for a major part of its life cycle in the extracellular space of a living plant . The specificity and extremely high affinity of this transporter enables the pathogen to compete efficiently and successfully with the adjacent cells of its host for sucrose at the plant/fungus interface ( Figure 8 ) . Srt1 is perfectly suited to out-compete both the plants sucrose transporters ( SUC or SUT proteins [27] ) with their comparatively low substrate affinities as well as the invertase ( INV ) -dependent plant monosaccharide transporter ( STP ) proteins that are thought to feed different plant sink tissues ( Figure 8 ) and that are known to be induced in response to elicitor treatment [36] or fungal infection [37] . Although most of STP proteins are high-affinity transporters , plant extracellular invertases have KM values in the millimolar range and , therefore , seem to represent the rate-limiting step [38] . Under growth chamber conditions , an U . maydis mutant that had its srt1 gene replaced by an srt1 promoter/AtSUC9 cDNA fusion showed wild-type virulence ( Figure 4A ) . With a KM-sucrose of 0 . 5 mM [32] , AtSUC9 has a lower substrate affinity than Srt1 , but still one of the lowest KM-sucrose values determined for plant sucrose transporters . In contrast , the KM-sucrose of ZmSUT1 , the sucrose transporter responsible for phloem loading in maize and , thus , the competing transporter at the U . maydis/maize interface , varies from 3 . 7 mM at pH 5 . 6 to 12 . 4 mM at pH 6 . 5 [39] . These different KM values may explain the successful replacement of Srt1 by AtSUC9 . Nevertheless , it could well be that SG200Δsrt1::AtSUC9 would show reduced virulence in the field , where growth conditions are more competitive . This result demonstrates that the primary function of Srt1 is , in deed , the supply of sucrose to the pathogen . Other possible functions , e . g . , the signaling by interaction with a protein partner can be excluded , as it is highly unlikely that a foreign protein , such as AtSUC9 , could complement such a function of Srt1 . Direct uptake of sucrose by a plant pathogenic fungus possibly provides also a second , more strategic advantage over the uptake of monosaccharides produced by the activity of a secreted fungal invertase . It was reported repeatedly that invertase-derived monosaccharides in the apoplast act as signaling molecules that trigger reduction of photosynthetic activity and induction of defense genes [13]–[16] , [40]–[42] . Both responses are highly unfavorable for a biotrophic pathogen , as the first would reduce carbon availability for the pathogen and the second could even stop the infection . The use of a sucrose transporter rather than of an invertase/hexose transporter pair might , therefore , represent a mechanism of signal avoidance in an environment that is well prepared to sense and destroy potential pathogens . The exclusive induction of srt1 expression in tumor tissue implies that the transporter is specifically employed for sucrose uptake at the plant/fungal interface . During saprophytic growth on sucrose containing media the gene is neither expressed nor needed , since Δsrt1 strains do not show reduced growth rates on media with sucrose as sole carbon source . As the presence of sucrose alone is not sufficient for srt1 induction ( Figure 3A and 3C ) , we must assume additional plant signals triggering the expression . Srt1 allows direct utilization of apoplastic sucrose without prior hydrolysis in the extracellular lumen . During evolution of pathogenicity , especially of biotrophic fungi , this may have been a major step to successfully adapt to the hostile environment in host plants . The extremely high sucrose affinity and specificity of Srt1 not only has advantages for the carbon acquisition of the pathogen . It also offers a mechanism to prevent plant defense responses by avoiding the production of signaling molecules in the plant apoplast .
Escherichia coli strain TOP10 ( Invitrogen ) was used for cloning purposes . For plant infections , U . maydis cells were grown at 28°C in YEPSL [43] . For RNA extraction , U . maydis was grown in glutamine minimal medium , which is based an the minimal medium described by Holliday [44] with 30 mM l-glutamine as nitrogen source . Plant infections with U . maydis were performed as described [45] . The U . maydis strain used in this study is SG200 , a haploid , solopathogenic strain that can infect maize plants without a mating partner [17] . S . cerevisiae strains used for analyses of Srt1 were EBY . VW4000 ( [25] MATa; leu2-3 , 112; ura3-52; trp1-289; his3-Δ1; MAL2-8c; SUC2; Δhxt1-17; Δgal2; Δstl1; Δagt1; Δmph2; Δmph3 ) , SEY2102 ( [26] MATα; ura3-52; leu2-3 , 112; his4-519; suc2-Δ9; gal2 ) , D458-1B ( [46] MATα; leu2; itr1; ino1 ) , and DBY2617 ( [31] MATa; his4-539; lys2-801; ura3-52; suc2-438 ) . Cells were grown in minimal medium ( 0 . 67% yeast nitrogen base without amino acids plus required amino acids depending on the strain ) containing 2% maltose ( EBY . VW4000 ) or glucose ( all other strains ) at 29°C . SG200Δsrt: the deletion of srt1 was performed by a PCR-based approach [47] . The promoter region of the srt1 was amplified by PCR using primers 2374_LB1 ( 5′-TGG CTG TCA AGC CTC TTG AAG CAG-3′ ) and 2374_LB2 ( 5′-GAT GGC CGC GTT GGC CGC CAT GGT TAA GAG CAA GGG CGA C-3′ ) , creating an SfiI site at the 3′-end . The 3′ UTR sequence was amplified using primers 2374_RB1 ( 5′-CAC GGC CTG AGT GGC CAT CTC ACC TGA AAC TCT GCA GGC G-3′ ) and 2374_RB2 ( 5′-GCG TGC TCA TGT AGA CGG GAT AGC-3′ , creating an SfiI site at the 5′-end . Both fragments were ligated to an SfiI HygR fragment [47] . The entire srt1 open reading frame ( ORF ) was replaced by a hygromycin resistance cassette in strain SG200 . SG200Δsrt::srt1-GFP was generated by fusing the ORF for eGFP to the 3′-end of the srt1 ORF deleting the srt1 stop codon . Primer pairs used to generate the flanks for homologous recombination were 2374_LB1Pf ( 5′-CGG GTC TCC CTT TCC TTC TTT TGC-3′ ) and 2374_LB2Pf ( 5′-GTT GGC CGC GTT GGC CGC TTG TGG ACT CGG CTG CAG AGT TTC-3′ ) for the flank matching the C-terminus of srt1 , and 2374_RB1Pf ( 5′-GTT GGC CTG AGT GGC CTT GCA CTG CAC ATT CAC TAG CGG C-3′ ) and 2374_RB2 ( 5′-GCG TGC TCA TGT AGA CGG GAT AGC-3′ ) for the flank matching the 3′ UTR . Primers 2374_LB2Pf and 2374_RB1Pf carry the SfiI sites compatible to eGFP cassette of pUMA317 containing the hygromycin resistance gene [48] . The eGFP construct was integrated into the native srt1 locus of SG200 by homologous recombination [47] . SG200Δsrt-srt1::ip: The Δsrt1 deletion strain was complemented with the srt1 gene under the control of its native promoter ( about 2 . 5 kb of upstream sequence ) three times independently by homologous recombination of pSRT1-GW into the ip-locus [43] . pSRT1-GW constructs were cloned according to the Gateway Cloning protocol ( Invitrogen ) . attB-flanked PCR products of the 4 . 2 kb srt1 locus were generated using primer pairs p2374_GW_for ( 5′-GGG GAC AAG TTT GTA CAA AAA AGC AGG CTG ACC ACC ATA AGT GCC ATT CTC GC-3′ ) and 2374Stop_GW_rev ( 5′-GGG GAC CAC TTT GTA CAA GAA AGC TGG GTT CAT TGT GGA CTC GGC TGC AGA GT-3′ ) . BP and LR reactions were performed in one-tube format reaction using p123-BB-GW1 as destination vector . p123-BB-GW1 is a derivative of p123 [49] , which was digested with the restriction enzymes HindIII and NotI , restriction sites were blunted using the Klenow polymerase , and the Reading Frame B Cassette was cloned into the plasmid backbone following the Gateway Vector Conversion System protocol . SG200Δsrt::AtSUC9: Promoter and 3′ UTR sequences of srt1 were amplified as described for the srt1 deletion constructs . Both fragments were ligated upstream ( promoter ) and downstream ( 3′ UTR ) of an SfiI 3xeGFP HygR fragment of pUMA647 ( K . Zarnack and M . Feldbrügge , unpublished data ) in a derivative of the TOPO cloning vector ( Invitrogen ) . A SfiI/AscI fragment containing the 3xeGFP ORF of pSRT3G was replaced by the AtSUC9 ORF that had been amplified with the primes AtSUC9c_SfiI_fwd ( 5′-GAG GCC AAC GCG GCC ACC ATG AGT GAC ATC CAA GCA AAA G-3′ ) and AtSUC9c_AscI_rev ( 5′-GGC GCG CCT TAA GGT AAA ACG GTA AGT GC-3′ ) that added SfiI and AscI cloning sites to the sequence . The resulting vector was pKW54 . To exchange the srt1 ORF of SG200 by AtSUC9 , pKW54 was linearized with KpnI and integrated by homologous recombination into the srt1 locus . The correct insertion was verified by Southern blot analysis of genomic DNA . Molecular methods followed described protocols [50] . DNA isolation from U . maydis and transformation procedures were performed as described [51] . Homologous integration of constructs was verified by gel blot analyses . Transformation of S . cerevisiae followed the protocol given in [52] . Total RNA from U . maydis cells grown in axenic culture was extracted using Trizol reagent ( Invitrogen ) according to the manufacturer's instructions . RNA samples to be used for real-time RT-PCR were further column purified ( RNeasy; Qiagen ) and the quality checked using a Bioanalyzer with an RNA 6000 Nano LabChip kit ( Agilent ) . The srt1 ORF was amplified from U . maydis genomic DNA using the primers 2374_EcoRI_for ( 5′-CAG AAT TCA AAA ATG GCG TCG TCT TCT CCC ATT CGT-3′ ) and 2374_EcoRI_rev ( 5′-CAG AAT TCT CGG ACT GCC AAG TCA TTG TGG AC-3′ ) . DNA was sequenced and cloned into the S . cerevisiae/E . coli shuttle vector NEV-E [53] , and the resulting plasmid was used for yeast transformation . For the fusion of Srt1 to the N-terminus of GFP , srt1 ORF was PCR-amplified with primers that removed the stop codon . The resulting srt1 ORF was cloned upstream of the ORF of GFP in the S . cerevisiae expression plasmid pEX-Tag [54] . S . cerevisiae cells were grown to an absorbance at 600 nm ( A600 nm ) of 1 . 0 , harvested , washed twice with water , and resuspended in buffer to an A600 nm of 10 . 0 . If not otherwise indicated , uptake experiments were performed in 50 mM Na-phosphate buffer ( pH 5 . 0 ) with an initial substrate concentration of 1 mM 14C-labeled sucrose ( or another 14C-labeled or 3H-labeled substrate ) . Cells were shaken in a rotary shaker at 29°C , and transport tests were started by adding labeled substrate . Samples were withdrawn at given intervals , filtered on nitrocellulose filters ( 0 . 8-µm pore size ) , and washed with an excess of distilled H2O . Incorporation of radioactivity was determined by scintillation counting . Competition analyses were performed with 0 . 1 mM 14C-sucrose in the presence of 10 mM competitor ( 100-fold excess ) . For analyses of the energy dependence of sucrose transport , d-glucose was added to the yeast cells 2 min before the start of the experiment to a final concentration of 10 mM . For inhibitor analyses , CCCP ( carbonylcyanide m-chlorophenylhydrazone ) or PCMBS ( p-chloromercuribencene sulfonate ) were used at final concentrations of 50 µM . For influx/efflux analyses in the plateau of sucrose accumulation ( Figure 6D ) , identical amounts of S . cerevisiae cells were incubated in two flasks with either 100 µM 14C-labeled sucrose or with unlabeled sucrose , and sucrose uptake was determined in the flask with the labeled substrate . When the plateau was reached ( after 35 min ) , the cells were quickly pelleted and washed in Na-phosphate buffer ( pH . 5 . 0 ) . Cells from the unlabeled flask were then resuspended to the initial volume with 100 µM 14C-sucrose , cells from the labeled flask with 100 µM unlabeled sucrose , and uptake experiments were continued . Light microscopic analyses were performed using a Zeiss Axioplan 2 microscope . Photomicrographs were obtained with an Axiocam HrM camera , and the images were processed with Axiovision ( Zeiss ) and Photoshop ( Adobe ) . Chlorazole Black E staining of fungal cells in planta was performed as described [55] . GFP signals of Srt1::GFP ( excitation at 450–490 nm , emission at 520 nm ) in infected plant tissue or in sterile cultures , and autofluorescence of plant cell walls ( excitation at 365 nm , emission at 397 nm ) were visualized using an Axio Imager ZI microscope ( Carl Zeiss ) . Images were processed with the AxioVision system ( Carl Zeiss ) . Subcellular localization of the Srt1::GFP fusion protein in S . cerevisiae was determined by confocal microscopy ( Leica TCS SPII; Leica Microsystems ) and processed with the Leica Confocal Software 2 . 5 ( Leica Microsystems ) . Emitted fluorescence was monitored at detection wavelengths longer than 510 nm . To analyze srt1 expression on different carbon sources , SG200 was grown in glutamine minimal media supplemented with the indicated amount of the respective carbon source to an optical density at 600 nm ( OD600 ) of 1 . 0 for 6 h . Precultures were grown overnight in glutamine minimal medium containing 1% of glucose . RNA samples were frozen in liquid nitrogen for two independently conducted replicates . RNA of maize plants infected with SG200 was prepared as described [45] . Samples were taken 0 . 5 , 1 , 2 , 4 , and 8 dpi . For cDNA synthesis , the SuperScript III first-strand synthesis SuperMix assay ( Invitrogen ) was used on 1 µg of total RNA . qRT-PCR was performed on a Bio-Rad iCycler using the Platinum SYBR Green qPCR SuperMix-UDG ( Invitrogen ) . The U . maydis actin ( um11232 ) and eIF2B ( um04869 ) genes were used as references . Primer sequences were rt-eIF-2B-F ( 5′-ATC CCG AAC AGC CCA AAC-3′ ) and rt-eIF-2B-R ( 5′-ATC GTC AAC CGC AAC CAC-3′ ) for eIF2B , rt-actin-F ( 5′-CAT GTA CGC CGG TAT CTC G-3′ ) and rt-actin-R ( 5′-CTC GGG AGG AGC AAC AAT C-3′ ) for the actin gene , and 2374_rt_for ( 5′-AGA CGC GTG GAA GGA CTT TCT TCG-3′ ) and 2374_rt_rev ( 5′-CCT AGC TCG AAC TTT GAC CAC CGC-3′ ) for srt1 . For the phylogenetic analysis of the U . maydis Major Facilitator Superfamily ( MFS ) and for the identification of the 19 members of the U . maydis sugar transporter superfamily , 86 amino acid sequences of putative MFS members were obtained at MUMDB ( IPR007114 Major facilitator superfamily; http://mips . helmholtz-muenchen . de/genre/proj/ustilago/ ) . Two sequences of U . maydis ammonium transporters were included as out-group ( Figure S1 and Table S1 ) . For comparative phylogenetic analysis of Srt1 , the amino acid sequence was aligned with 95 transporter sequences obtained by BLASTP analysis . This includes fungal and plant sequences with the highest similarity to Srt1 , fungal and plant sequences with highest homology to A . thaliana sucrose transporters , as well as fungal and plant ammonium transporter sequences as out-group ( Figure S3 and Table S2 . ) . Sequences were aligned with MAFFT version 6 using the global alignment G-INS-i . A phylogenetic tree was calculated using the minimum linkage clustering method ( http://align . bmr . kyushu-u . ac . jp/mafft/online/server/ ) . TreeIllustrator 1 . 0 . 1 was used to visualize the Nexus formats of the MAFFT results . | The plant parasitic fungus Ustilago maydis is a biotrophic pathogen that depends on live plant tissue for development . It is highly adapted to maize ( Zea mays ) , where it causes the corn smut disease . Fungal cells growing within the plant apoplast are surrounded by the host plasma membrane at all growth stages , thereby establishing tight interaction zones with the host cells that assure optimal access to host-derived nutrients , including organic carbon sources . Here , we focus on the previously unknown feeding mechanisms of this plant pathogen within its host plant . We identified a fungal plasma membrane transporter , Srt1 , that is expressed exclusively after plant infection and that turns out to be essential for virulence development of Ustilago in infected plants . Srt1 is the first characterized fungal transporter that allows direct utilization of sucrose without extracellular hydrolysis into monosaccharides , the carbon form more commonly taken up by pathogenic fungi . It is highly specific for sucrose , and its affinity largely exceeds that of equivalent plant transporters . This not only provides advantages for the carbon acquisition by the pathogen , but quite likely also offers a mechanism to prevent induction of plant defense responses known to occur upon apoplastic sucrose hydrolysis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"biology/plant-biotic",
"interactions",
"plant",
"biology/agricultural",
"biotechnology",
"plant",
"biology"
] | 2010 | A Novel High-Affinity Sucrose Transporter Is Required for Virulence of the Plant Pathogen Ustilago maydis |
Replicating chromatin involves disruption of histone-DNA contacts and subsequent reassembly of maternal histones on the new daughter genomes . In bulk , maternal histones are randomly segregated to the two daughters , but little is known about the fine details of this process: do maternal histones re-assemble at preferred locations or close to their original loci ? Here , we use a recently developed method for swapping epitope tags to measure the disposition of ancestral histone H3 across the yeast genome over six generations . We find that ancestral H3 is preferentially retained at the 5′ ends of most genes , with strongest retention at long , poorly transcribed genes . We recapitulate these observations with a quantitative model in which the majority of maternal histones are reincorporated within 400 bp of their pre-replication locus during replication , with replication-independent replacement and transcription-related retrograde nucleosome movement shaping the resulting distributions of ancestral histones . We find a key role for Topoisomerase I in retrograde histone movement during transcription , and we find that loss of Chromatin Assembly Factor-1 affects replication-independent turnover . Together , these results show that specific loci are enriched for histone proteins first synthesized several generations beforehand , and that maternal histones re-associate close to their original locations on daughter genomes after replication . Our findings further suggest that accumulation of ancestral histones could play a role in shaping histone modification patterns .
In addition to the information encoded in DNA sequence , replicating cells can inherit epigenetic information , which refers to variable phenotypes that are heritable without an underlying change in DNA sequence . It is widely accepted that chromatin , the nucleoprotein packaging state of eukaryotic genomes , provides one potential carrier of epigenetic information . Although definitive proof that chromatin per se carries epigenetic information during replication exists in very few cases [1] , genetic studies in numerous organisms have identified key roles for chromatin regulators in multiple epigenetic inheritance paradigms [2] , [3] . The idea that chromatin structure carries epigenetic information poses a central mechanistic question—since chromosome replication involves dramatic perturbations to chromatin structure ranging from old histone displacement to widespread incorporation of newly synthesized histones , how can chromatin states be stably maintained ? To understand the mechanism by which chromatin states could be inherited , it is necessary to understand the unique challenges posed by histone protein dynamics during replication [4]–[7] . First , histones must at least transiently dissociate from the genome during passage of the replication fork—if old histones carrying epigenetic information do not re-associate with daughter genomes at the location from which they came , this could lead to “epimutation , ” analogous to DNA bases moving relative to one another during genomic replication . Second , it is unknown to what extent newly synthesized histones deposited at different loci differ in their covalent modification patterns . Finally , how old histones influence new histones , the basis for positive feedback , can be considered analogous to asking what the equivalent of base-pairing is during chromatin replication . Classic radioactive pulse-chase studies demonstrated that , in bulk , maternal histones segregate equally to the two daughter cells [4] , [8]–[10] . It is unknown , however , whether maternal histones remain close to the locus from which they were evicted by the replication fork or whether maternal histones are incorporated at preferred genomic loci in the two daughter genomes [5] , [7] , [11] . The extent of maternal histone dispersal affects the stability of epigenetic states in theoretical models of chromatin inheritance [12] , making experimental determination of this parameter a key goal for epigenetics research . To address these fundamental questions , we carried out a genetic pulse-chase with epitope-tagged histone H3 [13] to follow ancestral H3 for several cell divisions after removal of the ancestral tag . We find that old histone proteins do not accumulate at epigenetically regulated loci such as the subtelomeres but instead accumulate at the 5′ ends of long , poorly transcribed genes . As expected , old histones do not accumulate at loci exhibiting rapid histone turnover , but we also find that 3′ to 5′ movement of old histones along coding regions and histone movement during replication are required to explain the patterns of ancestral histone retention we observe . We estimate that maternal histones stay within ∼400 bp of their original location during replication , providing the first measure of this crucial parameter . Finally , we identify a number of factors that affect old histone localization , such as topoisomerase I and the H4 N-terminal tail , which both affect the 5′ bias in localization patterns . In contrast , CAF-1 mostly affects histone turnover at promoters . Together , these results provide a detailed overview of the movement of ancestral histones across multiple cell generations and identify a number of mechanisms that play a role in shaping the landscape of ancestral histone retention .
To extend our analyses to the entire genome , we carried out deep sequencing of HA and T7 libraries . HA- and T7-tagged H3 were immunoprecipitated after the tag swap but before release from arrest ( 0 generations ) , after release into a G2/M cell cycle block , and at 1 , 3 , and 6 generations after release . Sequencing reads were mapped to the yeast genome , normalized for read count , and HA/T7 ratios were computed genome-wide . These data correlated well with our microarray data , and we further validated these measurements by q-PCR at SPA2 and BUD3 , two genes which both exhibit high and low HA/T7 ratios at their 5′ and 3′ ends , respectively ( Figure S3 ) . In previous work , we and others [13] , [15]–[17] , [21]–[23] showed that there is a partial correlation between transcription levels and replication-independent histone dynamics . To understand how transcription might affect multigenerational histone retention in our system , we aligned all yeast genes by their transcription start site ( TSS ) and clustered genes ( K-means , K = 5 ) based on the pattern of the 3-generation HA/T7 ratios along the gene body ( Figures S4 , S5 , Table S1 ) . We observed a striking enrichment of H3-HA just downstream of the 5′ ends of genes ( typically peaking around the +3 nucleosome ) . One exception to the 5′ pattern described is found in one cluster of short genes with uniformly low H3-HA levels ( Figure S4 , Cluster 1 ) , which is enriched for GO categories ( such as protein translation ) related to high gene expression levels . In contrast , long genes were generally associated with higher levels of ancestral H3 ( see for example Cluster 5 ) . To better visualize these trends , we sorted genes by the extent of ancestral H3 retention after 3 generations ( Figure 2A–B ) . Retention of ancestral histones correlates both with low expression levels and with longer genes ( Figure 2C–D , Figure S6 ) . While it is the case that longer genes tend to be expressed at lower levels than short genes ( Figure 2E ) , these factors are partially independent here—even when we focus on genes of 1–2 kb length , we still observe the correlations between ancestral histone retention and low expression ( Figure 2E–F , and see below ) . Interestingly , in both microarray and sequencing datasets we found that epigenetically repressed loci such as the silent mating loci and subtelomeres [24] , [25] did not preferentially accumulate ancestral histone proteins ( Figure 1C , Figure S7 ) —analysis of both unique and repetitive subtelomeric genes showed similar H3-HA retention patterns to euchromatic genes of similar length and expression . This was not a consequence of silencing defects in our strains , as they showed efficient mating ( unpublished data ) . What properties of short or highly transcribed genes might lead to loss of ancestral histones ? Replication-independent histone replacement occurs most rapidly over intergenic regions and over the coding regions of highly transcribed genes [15] , [17] , [21] , [23] , the converse of the pattern of ancestral H3 retention we observe . Indeed , ancestral histone retention is broadly correlated with “cold” regions of low H3/H4 turnover ( Figure 3A ) . Importantly , however , for a given level of H3/H4 turnover , ancestral H3 retention varied significantly—retention at a given nucleosome was better correlated with the average turnover rate of several surrounding nucleosomes than with the immediate turnover rate ( see , for example , Figure 3B–C ) . This observation suggests that maternal histones preferentially re-associate with daughter genomes near the location from which they originated—if old histones scattered randomly at replication , ancestral H3 retention patterns should more precisely anticorrelate with replication-independent turnover patterns , as is discussed in more detail below . Why do old histone proteins accumulate near the 5′ ends of genes ? We considered two alternative possibilities for classes of mechanisms causing this pattern . In the first mechanism , we reason that if histone proteins tend to maintain their locations along the genome , the 5′ enrichment of old histones implies that old 3′ histones are evicted and replaced by new histones during some phase of the cell cycle . However , previous measures of turnover in G1- or G2/M-arrested yeast [15] , [26] cannot explain the 5′/3′ ratios we observe . Furthermore , we found that mutations in candidate 5′/3′-marking complexes such as cohesin [27] , [28] or H3K4/K36 methylases [29] did not affect 5′-biased retention of old histones at target loci ( Figure S8A ) . A second possible explanation for widespread 5′ accumulation of ancestral histone proteins is that the histone proteins move from 3′ to 5′ over genes over time . This could result from RNA polymerase passage , because some RNA polymerases pass histone octamers in a retrograde direction during transcription [30] , [31] . Although it is debatable whether this is true of Pol2 in vitro [32] , [33] , in vivo we previously observed that inactivation of Pol2 leads to a modest shift of nucleosomes from 5′ to 3′ [34] , consistent with the idea that Pol2 movement normally shifts nucleosomes in a 5′ direction . To test whether this movement was related to transcription , we asked whether the 5′ peak of H3-HA accumulation shifted further 5′ with increasing transcription rate . We normalized all gene lengths to one , then plotted the HA/T7 ratio for all genes sorted by transcription rate ( Figure S9 ) . Consistent with the prediction of transcription-dependent retrograde movement , we did observe a subtle signal of H3-HA peaks shifting further 5′ at higher transcription rates . While this analysis could be confounded by the higher transcription rates seen over shorter genes , even when we focus on 1–2 kb genes , we observe that poorly transcribed genes exhibit a much flatter profile than genes expressed at average levels ( Figure 2F ) , as expected if Pol2 transit were required for H3/H4 “passback . ” Finally , we show below that per-gene estimates of passback exhibit significant correlation with Pol2 levels . Together , these results are most consistent with a model in which histone proteins move from 3′ to 5′ over coding regions over time ( further detailed in the Discussion ) . A key question we sought to address in this study is whether maternal histones re-associate near their original positions after passage of the replication fork . We reasoned that changes in HA/T7 patterns over the course of several generations might provide insight into the effects of replication on nucleosome dynamics . HA/T7 patterns change dramatically between arrest and 1 generation of release ( with or without G2/M arrest ) and then are very similar between 1 and 3 generations , before the background of nonswitching cells starts to dominate the profile at 6 generations ( Figure 4 ) . As expected , HA/T7 data at generation 0 exhibited widespread HA loss/T7 gain at promoters and +1 nucleosomes as a result of the rapid replication-independent turnover at these loci [13] , [15] , [17] . Importantly , to rule out the possibility that 5′ accumulation of H3-HA was an effect of our arrest-release protocol , we also measured HA/T7 distributions 6 h after inducing recombination in actively growing midlog cultures of yeast ( Figures S3 and S10 ) . Despite heterogeneity in switching times in this protocol ( only 65% of yeast have switched from HA to T7 3 h after switch induction , 85% after 6 h ) , we nonetheless observed that HA/T7 distributions were remarkably similar in midlog-switching cells to HA/T7 patterns observed in cells undergoing the arrest/release protocol , with preferential ancestral histone accumulation at the 5′ ends of long , poorly transcribed genes . We asked whether these dynamic observations could be used to quantitatively rule out specific models concerning the mechanisms for segregation of maternal histones to daughter genomes . However , the resolution of this question is complicated by replication-independent processes we discuss above that can remove or shift ancestral histones , and that cannot be fully removed experimentally ( for example , yeast will not proceed through the cell cycle in the absence of RNA polymerase ) . To understand the relationship between these issues , we designed an analytical model that accounts for three processes that affect H3 molecules in coding sequences ( Figure 5A ) and then examined the effect of removing any of the three . Briefly , our model includes a nucleosome-specific term for H3 turnover taken from prior experimental results [15] , with H3 turnover resulting in loss of HA . In addition , it includes a gene-specific parameter accounting for lateral movement of histones ( “passback” ) . Further , the model also includes a global parameter that describes the extent of histone “spreading” via dissociation/re-association during replication . Finally , the experimentally measured background of 2% nonswitching cells ( Figure S1 ) was included . The free parameters of the model ( describing global histone spreading and gene-specific lateral movement per generation ) were estimated to maximize the likelihood of experimental observations ( Text S1 ) . To account for any first-pass effects of Pol2 behavior during initial re-feeding of nutrient-depleted yeast ( Figures S3 and S10 ) , we examined this model with two starting conditions—the first started with a uniform genomic distribution of H3-HA , while the second started with the experimental distribution of HA/T7 observed after release into G2/M arrest ( Figure 4 ) . Both model variants predicted HA/T7 ratios with good correlations to the experimental data ( Figure 5B shows data starting from a uniform distribution , Figure 5E and Figure S11 start from the G2/M distribution ) . Examination of estimated parameters revealed expected behaviors . For instance , the distribution of lateral histone movement estimates ( Figure 5C ) was strongly biased towards retrograde 3′ to 5′ movement of histones , consistent with the previously measured effects of rpb1-1 inactivation on nucleosome positioning [34] . Passback values were also significantly ( p = 9 . 6439e-19 ) correlated with transcription rate ( Figure S12 ) . Our model allows us to estimate the extent of histone movement during replication . Figure 5D shows the likelihood of the full model plotted for various values for replication-dependent histone spreading . The best fit model allowed histones to spread ∼400 bp in either direction , or roughly two nucleosome widths , during replication ( more precisely , in this model two-thirds of histones stay within 400 bp of their original locations , as this value is the standard deviation of a Gaussian function describing spreading; see Text S1 ) . Results from models with 800 and 1 , 600 bp spreading parameters are shown in Figure 6 for comparison . Our estimate of ±400 bp spreading is particularly interesting given electron microscopy results demonstrating that nucleosomes are destabilized over 650–1 , 100 bp around the replication fork on replicating SV40 minichromosomes [35] , [36] . Elimination of any one or two of the three components of the model—spreading , turnover , or passback—resulted in significantly worse fits between model predictions and experimental data ( Figure 5E ) . This can be intuited as follows . First , in the absence of histone spreading , unmitigated histone movement from 3′ to 5′ results in a much tighter 5′ ancestral histone peak and results in much more extensive change from one generation to the next than we observe . Second , eliminating histone turnover shifts the 5′ ancestral peak closer to the +1/+2 nucleosome . Third , preventing lateral histone movement results in a 3′-shifted , flatter ancestral histone profile . While our model provided good quantitative fits of ancestral H3 patterns for many genes , we nonetheless note that many genes were not perfectly fit by this model . Generally , we found that the model poorly fit short genes , and overall the model almost universally predicted lower HA/T7 at the +N nucleosome ( the last nucleosome in a gene ) than was observed ( Figures S11 , S13 ) . We ascribe these failures to the fact that we considered each gene in isolation and therefore did not model shifts of old nucleosomes from adjacent genes , which would result in poor fits over short genes in particular . Interestingly , the better fit at the +1 nucleosome than at the +N nucleosome is consistent with rapid promoter turnover more effectively isolating genes from one another at their 5′ ends in vivo . Overall , the strong correlation between our model and the experimental data supports the hypothesis that at least three dynamic processes affect nucleosomes and shape the landscape of ancestral histone retention and provide the first quantitative estimate of maternal histone dynamics during replication . To further investigate the mechanism of 5′ accumulation , we asked whether gene-specific passback parameters were correlated with specific gene annotations ( Table S2 ) [34] , [37] . Interestingly , we find that the estimated passback distance was much greater at TFIID-dominated ( “growth” ) genes than at SAGA-dominated ( “stress” ) genes ( Figure 7A ) [38] . As a result , 5′ accumulation was much more pronounced at TFIID-dominated than at SAGA-dominated genes ( Figure 7B ) . Almost every described aspect of chromatin structure and gene expression , from nucleosome positioning to evolutionary lability ( reviewed in [39] , [40] , [41] ) , differs between these two broad types of genes . Mechanistically , one interesting correlate is that TFIID recruitment has been proposed to be mediated in part by acetylation of the N-terminal tail of histone H4 [38] , [42] . To investigate this link experimentally , we examined whether mutations of the H4 tail influenced ancestral histone H3 retention . In an H4K5 , 12R mutant that cannot be acetylated on these two tail residues , the 5′-biased HA/T7 was partially lost ( Figure S8B ) , consistent with the possibility that acetylation of H4 tail lysines may contribute to H3/H4 passback . We also deleted the H4 N-terminal tail , although in this strain background this mutation proved lethal and so all recovered strains retained a wild-type copy of the H4-H3 locus ( HHF2-HHT2 ) . Thus , results with this strain must be interpreted with extreme caution , as we do not know the effect of wild-type , untagged nucleosomes on the behavior of the epitope-tagged histones . Nonetheless we present here results of mapping of HA and T7 3 generations after release from the HA/T7 tag swap , since the H4 tail deletion has dramatic effects on global nucleosome dynamics ( Figure 7C ) , with low HA/T7 at 5′ ends followed by a nearly flat profile over the remainder of coding regions . This profile suggests a requirement for the H4 tail in H3/H4 passback , and possibly on replication-mediated spreading ( see Figure 5E ) . Interestingly , the effect of H4 tail deletion was much more pronounced at TFIID-dominated genes ( Figure 7D ) , suggesting that the exaggerated H3/H4 passback inferred at these genes involves the H4 tail . The effects of the H4 tail deletion were not simply due to the extensive changes in the transcriptome [43] , as we measured changes in genome-wide RNA Pol2 localization in our H4 mutant strains , finding that the relationship between Pol2 levels and HA/T7 behavior qualitatively changed in this mutant ( Figure S14 ) . While we must be cautious interpreting results obtained with the H4 tail deletion , the fact that H4K5 , 12R mutants ( which were viable and did not retain any wild-type H3/H4 ) also exhibit diminished 5′ bias in ancestral H3 retention provides independent support for a key role for the H4 tail in H3/H4 passback . We also explored the role of supercoiling in the 5′-biased retention of old histones . Topoisomerases relax DNA supercoiling and thereby help to maintain chromatin architecture . Transcription of DNA templates by Pol2 differentially affects supercoiling in front of and behind the passing polymerase , thereby differentially affecting 5′ and 3′ nucleosomes [44] , [45] . To assess the role of this activity in 5′ accumulation of old histones , we examined the consequences of inactivation of the major topoisomerase Top1 , which in vitro can resolve both negative and positive supercoils [46] , [47] . Cells lacking Top1 showed reduced 5′ bias in ancestral nucleosome accumulation ( Figure 7E ) , indicating that resolving DNA topology problems before or after passage of the transcription or replication machinery influences the mobility and/or stability of nucleosomes . Consistent with expectations of a greater buildup of supercoils over longer transcription units , we confirmed a stronger effect of TOP1 deletion at longer genes ( Figure 7F ) . We finally turn to the role of replication factors in ancestral histone retention . We first asked whether replication timing affected H3-HA retention . Nucleosomes surrounding early-firing origins tended to lose H3-HA more rapidly than late-firing origins ( unpublished data ) , but this likely stems from the fact that replication timing correlates with replication-independent turnover [23] , [26] . Focusing only on nearby coding regions ( Figure S15 ) , we found that late-replicating genes were associated with slightly 5′ shifted ancestral H3 peaks relative to genes near early origins ( consistent with decreased spreading or turnover ) , suggesting that different replication forks might affect chromatin in different ways , although the modest effect precludes a stronger interpretation . To directly address the role of fork-associated chromatin proteins in histone spreading at replication , we examined mutations of PCNA and Chromatin Assembly Factor ( CAF-1 ) , which plays a key role in replication-coupled histone deposition [48] , [49] . Three different mutants of PCNA that disrupt interactions with replication proteins or with replication-coupled chromatin-assembly factors showed only minor effects on 5′ retention of ancestral H3 at target genes SPA2 and BUD3 ( Figure S8C ) . In contrast , ancestral H3 retention at the 5′ ends of these target genes was slightly increased upon deletion of the CAF-1 subunit CAC1 ( unpublished data ) . To further explore the role of CAF-1 in histone retention patterns , we deep sequenced HA and T7 tags from cac1Δ yeast 3 generations after release ( Figure 8 ) . These data show a dramatic 5′ shift in the peak of ancestral H3 retention in these mutants . This shift is most consistent with a decrease in histone turnover at the 5′ ends of genes in this mutant , which we have independently confirmed using G1-arrested yeast expressing pGAL-driven Flag-H3 [50] . However , we cannot rule out the possibility that the role of CAF-1 in retention of old histones in 5′ and promoter regions involves interactions with PCNA during DNA replication . Interestingly , the 5′ accumulation observed in wild-type yeast is otherwise little changed in the CAF-1 mutant over long genes ( Figure S16 ) , suggesting that 3′ to 5′ movement of histones is normal , and that preferential retention of old histones at their maternal locations may be carried out by alternative histone chaperones such as the Hir complex or Asf1 in this mutant . Unfortunately , both hir and asf1 mutants are lethal in our strain background ( likely because our strain carries only one copy of the H3/H4 gene pair [51] ) , preventing us from testing this hypothesis . Our results are most consistent with histone retrogression from 3′ to 5′ over genes , which raises the question of whether old histones carry modifications associated with mid- and 3′ coding regions ( e . g . , H3K36 and H3K79 methylation ) towards the 5′ end of genes . Alternatively , there could be active erasure of these modifications . We therefore compared genes exhibiting high levels of ancestral H3 retention with prior genome-wide analyses of histone modifications [52] , [53] . Histone modification patterns generally conformed to the patterns expected based on transcriptional behavior—genes that retain high levels of ancestral histones are poorly transcribed ( Figure 2D ) , and correspondingly exhibit low levels of transcription-related marks H3K9ac , H3K14ac , H4ac , and H3K4 methylation ( Figure S17 and unpublished data ) . However , these are all 5′-biased marks [19] , [29] , [52] , [54] , and based on retrograde movement of old histones are therefore not expected to accumulate with age . More interestingly , we found that genes with high levels of old nucleosomes were enriched for H3K79me3 throughout their coding regions , particularly at the 5′ end ( Figure 9 , Figure S17 ) . The H3K79 methylase is nonprocessive , indicating that K79 methylation status should essentially act as a timer [55] . Further , analyses of genome-wide H3K79me3 patterns show anticorrelation between this modification and locations of high nucleosome turnover [15] , [56] , supporting the idea that K79me3 identifies old H3 protein . We recently confirmed that old H3 protein is enriched for H3K79me3 by mass spec analysis of old nucleosomes ( D . DeVos , FvL et al . , submitted ) . Finally , we also observed higher levels of H3K36me3 at 5′ and mid-CDS of genes exhibiting elevated ancestral histone retention relative to genes with intermediate H3-HA retention ( Figure S17 ) . This observation is consistent with the above hypotheses that old histones move from 3′ to 5′ and thus might carry typical mid-CDS and 3′-end histone modifications to the 5′ ends of genes ( Figure S18 ) . Together , these results provide further evidence that our system accurately captures the behavior of old histones .
Most surprising to us was the observation that ancestral H3 molecules accumulate near the 5′ ends of coding regions , peaking around the +3 nucleosome . The high HA/T7 ratio observed at the 5′ ends of genes is not an artifact of the epitope tags used , as we have observed the converse behavior ( high T7/HA ) when we switch the epitope tags used ( unpublished data ) . Furthermore , this unusual behavior is not an artifact of the conditions used for growth arrest and release , as we observe a similar 5′/3′ gradient of H3-HA when yeast are subjected to the epitope switch during active midlog growth ( Figure S10 ) . What is the mechanistic basis for the 5′/3′ gradient of HA/T7 we observe ? We consider two classes of mechanisms—in one , histone proteins do not move laterally and the 5′/3′ gradient results from preferential loss of 3′ H3/H4 , while in the other the gradient results from lateral histone movement combined with loss at the 5′ end . While we cannot definitively answer which mechanism explains our results , we strongly disfavor a model with preferential 3′ nucleosome eviction and no lateral movement based on the following observations . First , we tested a number of relevant mutants for changes in the 5′/3′ HA/T7 bias ( Figure S8 ) . Loss of H3K4 methylation ( a 5′-biased histone mark ) or H3K36 methylation ( a mid and 3′-biased histone mark ) did not affect HA/T7 patterns at selected target genes . Similarly , 5′ retention of ancestral H3 was unaffected by mutants of cohesin , whose loading is associated with regions of high H3/H4 turnover and which accumulates at the 3′ ends of genes [27] , [28] . Second , direct measurements of H3/H4 turnover using a pGAL-driven epitope tagged H3 do not provide evidence for ubiquitous 3′ histone replacement during G1 arrest [15] , during G2/M arrest [26] , or in unsynchronized yeast [15] . Thus , while we cannot definitively rule out some cryptic 3′ replacement event in this system , all direct tests have failed to support this hypothesis . Conversely , multiple observations support the hypothesis that H3/H4 proteins move from 3′ to 5′ over protein-coding regions over time . First , seminal in vitro studies on transcription of nucleosomal templates showed that several RNA polymerases can transcribe through a nucleosome without displacing the H3/H4 tetramer . The proposed mechanism by which histones remain associated with the DNA is a “bubble propagation” mechanism—DNA partially unwraps from the histones , RNA polymerase enters , and DNA behind the polymerase re-associates with the histone octamer , resulting in a net retrograde movement of histones after the polymerase has passed . This mechanism is relatively well established for SP6 polymerase and RNA Polymerase III [30] , [31] , whereas there is some controversy regarding the effect of RNA Polymerase II on nucleosome positioning [32] , [33] . Of course , it is not unreasonable to expect that nucleosome movement during transcription in vivo will also be affected by polymerase-associated factors such as histone chaperones and ATP-dependent remodelers that are not present in the in vitro systems . In any case , these studies provide a plausible mechanism by which RNA polymerase transit results in retrograde nucleosome movement . Second , we have previously found that inactivation of Pol2 using the temperature-sensitive rpb1-1 allele results in a net 5′ to 3′ shift in the majority of coding region nucleosomes [34] , consistent with the hypothesis that polymerase transit normally shuttles nucleosomes from 3′ to 5′ . Third , highly transcribed genes ( such as those encoding ribosomal proteins ) in yeast paradoxically exhibit very tightly spaced coding region nucleosomes ( e . g . , 155–160 bp between adjacent nucleosomes rather than ∼165 bp ) , and this tight spacing relaxes upon Pol2 inactivation , again consistent with nucleosomes being passed upstream during transcription [34] . Taken together with the absence of any evidence for 3′ H3/H4 eviction , we therefore argue that the most parsimonious explanation of the surprising 5′ accumulation of ancestral histones is retrograde movement of histones over genes against the direction of transcription . Note that while we favor the hypothesis that the act of RNA polymerase transit itself is the mechanism linking transcription to H3/H4 passback , polymerase is not the only candidate factor leading to retrograde histone movement . Notably , we found that top1Δ mutants exhibit diminished signatures of H3/H4 passback ( Figure 7 ) , and this decrease was stronger at longer genes , suggesting the possibility that some aspect of cleavage and rotation of twisted DNA by Top1 contributes to the passback observed . However , it is also possible that Top1 differentially affects histone turnover in 5′ and 3′ regions or affects passback by affecting Pol2 passage [60] . We analytically assess several predictions of the “passback” model . First of all , if RNA polymerase transit were the driver of retrograde histone movement , then one might predict that passback should correlate with transcription rate . We find the expected correlation to be statistically significant ( p = 9 . 6439e-19 , Figure S12 ) but weak nonetheless ( R = 0 . 12 ) . Importantly , we previously observed that 5′ to 3′ nucleosome movement in rpb1-1 mutants was also significantly but poorly correlated with transcription rate [34] . The reason for the mediocre correlation between polymerase abundance and passback is hinted at by the fact that TFIID-dominated genes exhibit much greater passback values than do SAGA-dominated genes ( Figure 7A ) . We have previously noted that SAGA-dominated ( “stress” ) genes exhibit higher levels of H3 turnover , per polymerase , than do TFIID-dominated genes [15] . In vitro , a single polymerase's transit displaces an H2A/H2B dimer from the histone octamer , but a second polymerase encountering a histone hexamer will displace the remaining histones [61] , [62] . Coupled with the observation that SAGA-dominated genes exhibit larger “bursts” of polymerase , this suggests that closely spaced polymerases are required for H3/H4 eviction over coding regions , but evenly spaced polymerases leave time for dimer replacement on damaged nucleosomes [54] , [62] . We believe this model also explains some of the behavior of ancestral histones in this study—SAGA-dominated genes display little passback and overall diminished levels of ancestral H3 ( Figure 7A–B ) , an expected consequence of the loss of old histones via turnover . Correlations between polymerase and passback are therefore expected to be subtle—at increasing transcription rates , we expect an increased likelihood of a closely spaced pair of polymerases , and the resulting H3/H4 eviction would eliminate any trace of the passback that had occurred to that point . It is important to note that the transcription-dependent passback postulated here cannot simply be interpreted as a model in which every round of polymerase passage shifts the histone octamer upstream by one position ( ∼165 bp ) . In Figure 7A , our estimates of passback per cell cycle have a mean of ∼90 bp at TFIID-dominated genes , less than the spacing between adjacent nucleosomes . If taken literally , these values would be difficult to reconcile with the observation that the majority of yeast nucleosomes are well positioned [20] . Instead , we interpret the passback values in terms of probability that an octamer will be passed back in a given cell cycle in each cell—a passback value of ∼80 bp suggests that there is a 50% chance that histones on a given gene will be shifted back one position towards the 5′ end in a single cell cycle . Physically , we imagine that polymerase passage results in relatively short retrograde movement of H3/H4 , which then have some probability of returning to their original position , and some probability of shifting to a new upstream location . Our results show a surprising pattern of ancestral histone retention in yeast , with old histone proteins accumulating near the 5′ ends of genes—the histone proteins located at the +3 nucleosome are the oldest histone proteins over a typical yeast gene . These data are best explained by a model in which H3/H4 proteins shuttle from 3′ to 5′ over coding regions over time , with eventual loss of old histone proteins when they are eventually moved into the +1 and +2 nucleosome positions . The process of genomic replication is enormously disruptive to chromatin structure , as the melting of the DNA double helix is accompanied by histone dissociation from the genome [4]–[7] . Thus , understanding where maternal histones re-associate relative to the locus from which they were evicted is a key constraint for understanding the potential of chromatin as an epigenetic information carrier . The ideal experiment for measuring this would be to epitope tag the histones at one specific locus ( e . g . , the +5 nucleosome over BUD3 ) in a large population of yeast , allow replication to proceed , and measure the new locations of the tagged histones . Despite numerous attempts , this type of tagging has proven technically intractable to date . Here we measure instead the bulk distribution of ancestral histones . Importantly , this still provides information on locus-specific histone behavior—as turnover rates are not homogeneous across the genome , even before we release yeast into the cell cycle the landscape of H3-HA exhibits variability ( Figure 4 , see generation 0 ) , and so in effect only a subset of ancestral locations are epitope-tagged before release . This enables us to infer the dynamic behavior of histone proteins during replication via analysis of the evolution of the H3-HA distributions over time . Two observations provide an intuition regarding the effects of replication on histone locations . First , ancestral histone retention exhibits the expected anticorrelation with replication-independent turnover ( Figure 3A ) . However , old histones are more efficiently retained at cold ( low turnover ) loci that occur in long cold domains , whereas short domains of cold nucleosomes lose ancestral histones over time . This observation is inconsistent with two extreme models for histone behavior during replication—if old histones were to completely dissociate from the genome during replication and randomly re-associate with the genome , then ancestral histone retention should precisely recapitulate turnover measurements . Conversely , if old histones were to reassociate precisely with their original locations , then ancestral retention should essentially integrate turnover for multiple generations . Thus , some process that shuffles histone proteins locally must be invoked along with turnover to shape the ancestral retention landscape . In principle , the preferential retention of old histones on longer genes could simply result from passback—shorter genes will more quickly have all of their histones passed “over a cliff” at the 5′ end . However , we find relatively static 5′/3′ gradients of old histone retention over time ( Figure 4 ) . While it is the case that H3-HA domains gradually shorten over time as predicted by the model that passback results in old histones being moved to promoters where they are replaced ( unpublished data ) , this effect is subtle and is quantitatively much less dramatic than predicted from passback alone . This leads to the second intuition regarding histone spreading during replication . Many examples exist for relatively static gradients in biological systems being established via a combination of directional active transport coupled with passive diffusion . Most relevant in our opinion is the “pump leak” model [63] for membrane ion gradients—active transport of ions across membranes , coupled with a passive leak of ions back into the cell , results in a static gradient . Here , we envision transcription-related passback as the active transport mechanism , with spreading during replication being somewhat analogous to the leak that results in a steady gradient rather than a continuous 3′ to 5′ march of histone proteins . We present a quantitative model that recapitulates our experimental data with only three dynamic processes—turnover , passback , and spreading . Locus-specific turnover rates were previously measured [15] and are not fit by the model . Passback is estimated for each gene separately , while spreading is a single global parameter affecting all histones . Thus , our model has 4 , 811 free parameters , which are used to fit over 100 , 000 HA/T7 ratios . This model does not overfit the data , and this can best be appreciated by the fact that eliminating a single parameter ( spreading ) greatly diminishes the agreement between model and data . Using this model , we estimate that maternal histones spread little ( ∼1–2 nucleosomes ) during replication . This value has not been measured before but is consistent with several related observations . First , electron microscopy studies on replicating chromatin show a stretch of ∼650–1 , 100 bp of nucleosome-free DNA surrounding replication forks [35] , [36] , consistent with histone movement of ±400 bp we estimate here . Second , histone proteins are retained in cis during in vitro replication even in the presence of competitor DNA [64]–[67] , indicating that histones do not freely diffuse away from replication forks but likely are retained locally . Finally , we previously observed that upon gene repression , loss of the active chromatin mark H3K4me3 occurs during S phase , but at very highly methylated nucleosomes H3K4me3 does not return to baseline levels immediately , with methylation levels falling little more than the 2-fold predicted by a dilution-based mechanism ( see Figure S5 in [68] ) . This final result indicates that “overmethylated” old histone proteins are retained near their original location , since extensive spreading of old histone proteins would enable a greater than 2-fold drop in methylation levels during S phase . Together , these results support the prospect of chromatin as a “sloppy” epigenetic information carrier ( “sloppy” in the sense that some spreading of histones will preclude mononucleosome-resolution information passage ) [69] , even if chromatin-based inheritance occurs infrequently [1] . Thus , chromatin states are unlikely to be inherited with mononucleosome precision , a view consistent with the fact that most or all proposed epigenetic chromatin domains are associated with long ( >1 kb ) blocks of histone modifications such as H3K9me3 or deacetylated H4K16 ( reviewed in [54] ) . To further investigate the mechanisms underlying the patterns of ancestral H3 retention , we assessed HA/T7 ratios at target genes in 12 mutants and further characterized HA/T7 genome wide for three of these mutants . Interestingly , a number of histone modifying factors , including Swd1 , Swd3 , Rtt109 , Nhp6 , and Set2 , had either no effect or subtle effects ( e . g . , Rtt109 ) on the 5′ accumulation of old histones at our target genes ( Figure S8 ) . These results suggest either that these mutants will have subtle global effects on ancestral H3 retention or that they have more localized roles that do not extend to the two target genes on which we focused . The three mutants we characterized at full genome coverage each had a distinct effect on ancestral H3 retention . Most dramatically , loss of the H4 N-terminal tail abolished the 5′ accumulation of ancestral histones—while the H4 tail deletion results are complicated by the retention of wild-type H3/H4 in this strain , the fact that similar results were obtained with clean H4K5 , 12R mutants ( Figure S8B ) provides independent support for observations obtained with the H4 tail deletion . The mechanistic basis for the loss of 5′ H3-HA retention is unknown to us—a flat HA/T7 profile is of course consistent with complete loss of passback . Alternatively , the observed profile in this mutant could be consistent with complete shuffling of maternal histones every generation , which as described above would be expected to more closely recapitulate a turnover-dominated profile . Importantly , loss of the H4 tail also affects H3/H4 turnover—in Figure 7D , the increased HA/T7 ratio at the 5′ ends of SAGA-dominated genes suggests a decrease in histone turnover in this mutant , and we have independently confirmed a decrease in replication-independent turnover in this mutant ( F . v . L . , manuscript in preparation ) . Analysis of Pol2 ChIP in H4 tail deletions shows that the effects of the H4 tail do not simply reflect altered transcription but instead reflect a change in the relationship between RNA Polymerase and histone dynamics over genes in this mutant ( Figure S14 ) . We also observe a similar , albeit muted , effect of Topoisomerase I on the 5′ accumulation of ancestral histones . Interestingly , loss of both topoisomerase I and II affects nucleosome occupancy and dynamics in S . pombe , indicating that topoisomerases play key roles in histone dynamics [60] . Here , we find that top1Δ mutants exhibit diminished 5′ accumulation of ancestral H3 and that this effect is stronger at longer genes than at shorter genes . As RNA polymerase passage will cause greater changes in supercoiling over longer genes , the preferential effects of Top1 on longer genes is consistent with the observation in S . pombe that topoisomerase mutants show evidence of stalled or slowed RNA polymerase over longer genes [60] . In addition to its role in transcription , topoisomerase I plays a key role in replication [45] . We note that the profile of top1Δ mutants here most closely mimics the predictions of our analytical model with both passback and spreading being compromised , but since neither of these is likely to be completely eliminated in top1Δ mutants , more detailed kinetic analyses will be required to make a quantitative statement about the role of Top1 in replication-related movement of histones . Finally , we assessed the role of the histone chaperone CAF-1 in ancestral H3 retention . To our surprise , we found that H3-HA exhibited even stronger 5′ accumulation in this mutant , with the 5′ peak of HA/T7 occurring closer to the +1 or +2 nucleosome ( compared to the +3 peak location for wild-type strains ) . This result most closely matches the predictions of a model in which H3/H4 turnover has been slowed without loss of passback or spreading ( Figure 5E ) . We recently tested this prediction using an alternative system for measuring replication-independent turnover ( pGAL-driven Flag-H3 ) and confirmed the prediction that caf mutants affect replication-independent histone replacement [50] . As CAF-1 and the Hir complex are known to complement one another in yeast , we predict that a caf hir double mutant would be necessary to uncover effects of replication-coupled spreading . Unfortunately , since both hir and asf mutants are lethal in our strain background , this prediction cannot be tested at present . Taken together , our results provide a surprising view of histone dynamics over multiple generations , with 5′ accumulation of ancestral histone proteins over coding regions and little evidence for preferential histone retention at epigenetically regulated loci such as subtelomeric genes . One unanticipated implication of this observation is that 3′ histone marks are expected to move towards the 5′ ends of genes over time , thereby shaping histone modification profiles ( as we document in Figure 9 and Figure S17 ) . This potentially necessitates mechanisms for erasure of these inappropriate marks in order to maintain accurate encoding of gene polarity . However , we note that active erasure of H3K4me3 after gene repression occurs most efficiently at 5′ ends of genes , whereas nucleosomes over coding regions mostly lose H3K4me3 by passive dilution ( [68] , see Figure S9 ) . If other old histone marks are not erased over coding regions , then we speculate that the accumulation of old histone proteins at the +3 nucleosome could potentially provide a mechanism by which a gene's transcriptional history could be integrated to play a role in regulation of the transition from transcriptional initiation to elongation . Most importantly , we find that old histones do not re-associate with daughter genomes at precisely the locus from which they dissociated . Thus , any inheritance of chromatin states must occur at the scale of ∼5–10 nucleosome domains rather than at single nucleosome resolution . These results therefore constrain the maximum amount of information theoretically carried by chromatin between generations . It will be of great interest in future studies to identify mutants that affect histone movement during replication and to measure their effects on the stability of epigenetic inheritance and to measure how maternal histone incorporation differs between leading and lagging strand daughter genomes .
For tag switch experiments , yeast cells were grown overnight in YPD in the presence of Hygromycin B ( 200 µg/mL , Invitrogen ) . The cells were then diluted 1∶10 into fresh YPD and incubated for 30–36 h . Recombination was induced by the addition of 1 µM β-estradiol ( E-8875 , Sigma-Aldrich ) . Subsequently , cells were diluted 1∶25 in fresh YPD media to release the cells back into the cell cycle and kept in log phase by 1∶2 dilutions into fresh media after each population doubling . Samples were taken after 1 , 2 , 3 , and 6 cell divisions or after 5 h of G2/M arrest . The number of population doublings was determined by microscopy and OD . G2/M arrest was induced by addition of 15 µg/ml Nocodazole ( Sigma-Aldrich ) and confirmed by FACS analysis . Strains are listed in Table S3 . Gene deletion mutants isogenic to strains NKI2048 , NKI2148 , and NKI2048 were made by homologous recombination using KanMX and/or NatMX selection markers . Gene deletion mutants isogenic to NKI4128 were made by crossing NKI4114 with gene deletion mutants from the MATa yeast knock-out collection using Synthetic Genetic Array methods . Histone mutants were made by transformation of strain NKI2148 with a HHF2-HHT2 CEN plasmid ( pMP9 ) , subsequent deletion of the tagged HHF2-HHT2 locus , followed by transformation with a PCR fragment encoding wild-type or mutated HHF2 in combination with tagged HHT2 . Deletion of the wild-type locus was confirmed in H4K5 , 12R mutants , whereas all surviving H4 tail deletion mutants retained a copy of the wild-type HHT2-HHF2 locus . ChIP was performed as described previously [13] , [70] with the following modifications . All steps were done at 4°C unless otherwise indicated . Following cell lysis by bead beating the insoluble chromatin of 1×109 cells was washed , resuspended in 400 µl FA lysis buffer ( 50 mM HEPES-KOH [pH 7 . 6] , 150 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% sodium deoxycholate ) , and sheared using a Bioruptor ( Diagenode ) for 6 min with 30 s intervals at high . The soluble fraction was diluted 3-fold in buffer 15 mM Tris-HCl pH 7 . 4 , 50 mM NaCl , 1 . 5 mM CaCl2 , 5 mM β-meracptoethanol , 5 mM MgCl2 , after which 25 units of micrococcal nuclease ( Worthington ) were added . The digestion reaction was incubated 20 min at 37°C and stopped by the addition of 10 mM EDTA and 10 mM EGTA; tubes were placed on ice . The majority of obtained fragments was around 150 bp , as determined on a 2% TAE agarose gel stained with ethidium bromide . The isolated chromatin of the equivalent of 3×108 cells was immunoprecipitated overnight at 4°C using magnetic Dynabeads ( Invitrogen ) , which were previously incubated with antibody O/N at 4°C . ChIP DNA was quantified by real-time quantitative PCR using the SYBR Green PCR Master Mix ( Applied Biosystems ) and the ABI PRISM 7500 . An input sample was used to make a standard curve , which was then used to calculate the IP samples , all performed in the 7500 fast system software . Primers used for qPCR are listed in Table S4 . ChIP DNA was treated with CIP ( calf alkaline phosphatase NEB; in 1× NEB buffer 3 , 0 . 25 U/µl CIP; 45 min at 37°C , reaction clean up with Qiagen MinElute spin columns ) . 20–150 ng of CIP treated ChIP DNA fragments were blunt ended and phosphorylated with the EPICENTRE End-it-Repair kit ( 1× buffer , 0 . 25 mM dNTPs , 1 mM ATP , 1 µl/50 µl reaction of Enzyme mix ) for 1 h at RT and cleaned up with Qiagen MinElute spin columns . Adenosine nucleotide overhangs were added using EPICENTRE exo-Klenow for 45 min at RT ( with 0 . 2 mM dATP ) . Illumina genome sequencing adaptors were then ligated using the EPICENTRE Fast-Link ligation kit: 11 . 5 µl A tailed DNA eluted from a MinElute column was mixed with 1 . 5 µl 10× ligation buffer , 0 . 75 µl 10 mM ATP , 0 . 5 µl Illumina DNA adaptors , and 1 µl Ligase . The reaction was incubated for 1 h at RT and subsequently supplemented with 7 . 5 µl water , 1 µl 10× buffer , 0 . 5 µl 10 mM ATP , and 1 µl ligase , and incubated overnight at 16°C . The ligation reaction was cleaned up with MinElute columns ( with an additional wash step to eliminate all the excess adaptors ) and the adaptor ligated fragments were amplified by PCR as follows: 0 . 5 µl of each Illumina genomic DNA sequencing primers , 10 µl 10× Pfx buffer 3 µl 10 mM dNTPs , 2 µl 50 mM MgSO4 , and 1 µl Pfx DNA polymerase ( Invitrogen ) were added to 30 µl DNA template in a 100 µl reaction . The cycling parameters were: ( 1 ) 94°C , 2′; ( 2 ) 94°C , 15″; ( 3 ) 65°C , 1′; ( 4 ) 68°C , 30″; ( 5 ) repeat from ( 2 ) 17 times; ( 6 ) 68°C , 5′ . The PCR product ( 200 to 300 bp in size ) was gel purified from a 2% TAE agarose gel using the Freeze'N Squeeze columns ( BioRad ) . Gel purified fragments were finally precipitated with Sodium acetate and Ethanol and pellets were resuspended ( 25 nM final concentration ) in TE buffer and sent for SOLEXA sequencing at the UMass Worcester core deep sequencing facility . Cells were grown as described above . Cell pellets ( ∼109 cells ) were flash frozen after formaldehyde crosslinking ( 1% ) and kept at −80°C overnight . Frozen cell pellets were resuspended in 300 µl cell braking buffer ( 100 mM Tris pH 7 . 9 , 20% glycerol , 1× Sigma Protease inhibitors cocktail ) and cell walls were broken down by bead beating using 400 µl of 0 . 5 mm zirconia/silica beads ( BioSpec Products ) in the BioSpec Mini-BeadBeater Model 8 three times for 1 min with 1 min pauses in between . Cell pellets ( 5 min max speed spin in refrigerated microcentrifuge ) were then washed once and resuspended in 800 µl FA lysis buffer ( with 1× Sigma Protease inhibitors cocktail ) . Chromatin was sheared by sonication in a cup sonicator ( Branson , 50% pulse at strength 7 for 3 . 5 min ) to 250–400 bp fragments . The sheared chromatin suspension was pre-cleared with 100 µl Protein A-agarose slurry ( IPA 400 HC RepliGen ) at 4°C for 1 h . 100 µl of the pre-cleared solution was saved for the ChIP input sample and 7 µl of RNA Pol II antibody ( abcam ab81859 , lot #: 933570 and GR6094-1 ) was added to the rest and incubated overnight at 4°C with rotation . ChIP DNA isolation and amplification by TLAD was done as described previously [19] . 2 . 5 μg of aRNA produced from the linear amplification were used to label probes via the amino-allyl method as described on www . microarrays . org . Labeled probes were hybridized onto a 4X44K yeast whole genome array ( Agilent ) at 65°C for 16 h . The arrays were scanned with the Agilent microarray scanner . | It is widely believed that chromatin , the nucleoprotein packaged state of eukaryotic genomes , can carry epigenetic information and thus transmit gene expression patterns to replicating cells . However , the inheritance of genomic packaging status is subject to mechanistic challenges that do not confront the inheritance of genomic DNA sequence . Most notably , histone proteins must at least transiently dissociate from the maternal genome during replication , and it is unknown whether or not maternal proteins re-associate with daughter genomes near the sequence they originally occupied on the maternal genome . Here , we use a novel method for tracking old proteins to determine where histone proteins accumulate after 1 , 3 , or 6 generations of growth in yeast . To our surprise , ancestral histones accumulate near the 5′ end of long , relatively inactive genes . Using a mathematical model , we show that our results can be explained by the combined effects of histone replacement , histone movement along genes from 3′ towards 5′ ends , and histone spreading during replication . Our results show that old histones do move but stay relatively close to their original location ( within around 400 base-pairs ) , which places important constraints on how chromatin could potentially carry epigenetic information . Our findings also suggest that accumulation of the ancestral histones that are inherited can influence histone modification patterns . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"genetics",
"biology",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | Patterns and Mechanisms of Ancestral Histone Protein Inheritance in
Budding Yeast |
The country of Kiribati is a small Pacific island nation which had a new case detection rate of 191 per 100 , 000 in 2016 , and is one of the few countries yet to reach the WHO leprosy elimination goal . Chemoprophylaxis of household contacts of new cases , or to the whole population in a highly endemic areas have been found to be effective in reducing new case rates . This study investigated the potential impact of different chemoprophylaxis strategies on future cases in South Tarawa , the main population centre of Kiribati . The microsimulation model SIMCOLEP was calibrated to simulate the South Tarawa population and past leprosy control activities , and replicate annual new cases from 1989 to 2016 . The impact of six different strategies for delivering one round of single dose rifampicin ( SDR ) chemoprophylaxis to household contacts of new cases and/or one or three rounds of SDR to the whole population was modelled from 2017 to 2030 . Our model predicted that continuing the existing control program of high levels of public awareness , passive case detection , and treatment with multidrug treatment would lead to a substantial reduction in cases but this was less effective than all modelled intervention scenarios . Mass chemoprophylaxis led to a faster initial decline in cases than household contact chemoprophylaxis alone , however the decline under the latter was sustained for longer . The greatest cumulative impact was for household contact chemoprophylaxis with three rounds of mass chemoprophylaxis at one-year intervals . The results suggest that control of leprosy would be achieved most rapidly with a combination of intensive population-based and household chemoprophylaxis . These findings may be generalisable to other countries where crowding places social contacts as well as household contacts of cases at risk of developing leprosy .
Despite being one of the most ancient diseases , there are annually still around 200 , 000 newly diagnosed leprosy cases worldwide [1] . Leprosy is one of the neglected tropical diseases , affecting the poorest and least-developed countries [2] . Most cases are in India , Brazil , and Indonesia , however 22 countries are classified as “high burden” by the World Health Organization ( WHO ) [3] . Kiribati , formerly known as the Gilbert Islands ( or Gilbert and Ellice Islands ) , is a country of 33 coral atolls and islands spanning an area of 3 . 5 million square kilometres of the Pacific Ocean . The widely dispersed population of 110 , 136 ( 2015 ) [4] provides challenges to delivering effective health care because of high transport and shipping costs , and limited communication infrastructure . Kiribati reached the WHO leprosy elimination goal of less than one case per 10 , 000 population in 2000 , however this was unable to be sustained . The new case detection rate ( NCDR ) in 2016 was 191 per 100 , 000 [1] . There is a relatively high proportion of paucibacillary ( PB ) disease , a low rate of grade-2-disability , and ongoing spread including to children [3] . The high NCDR has led the Ministry of Health and Medical Services ( MHMS ) in Kiribati to identify control of leprosy as a priority in their national development plan [5] . To reduce the high rates , the Kiribati MHMS has partnered with the Pacific Leprosy Foundation ( PLF ) to improve several aspects of the programme , including resourcing , database management , diagnostic skills , intensive case finding , improved publicity campaigns , and follow-up of patients and contact tracing . This has led to an increase in reported cases in recent years . South Tarawa is the political and economic centre of the country . Along with Betio , an islet connected to South Tarawa by a causeway , South Tarawa has experienced major population expansion through high birth rates and internal migration . Currently about half the population lives there , but the very limited land for settlement has caused increasing crowding in the heavily populated areas , with population densities in 2015 of 2 , 772 people per km2 in South Tarawa , and 10 , 377 people per km2 in Betio [4] . Land constraints have been exacerbated by the threat of sea level rise from climate change . Chemoprophylaxis aims to prevent the development of symptomatic disease in those with subclinical disease . Early research interest in chemoprophylaxis for leprosy waned with the introduction of multidrug therapy ( MDT ) , but has recently increased again . In 2005 Bakker et al . found a 75% reduction in new cases when two doses of rifampicin was given to the entire population of an isolated highly leprosy-endemic Indonesian island , while chemoprophylaxis of household contacts was not effective in this setting [6] . In 2008 the COLEP trials investigated the impact of single dose rifampicin ( SDR ) in spatially and genetically defined contacts [7] . They found an average reduction of 57% after two years . Both studies found that chemoprophylaxis was most effective in more distant contacts , likely because their subclinical infection is at an earlier stage , having had less intense exposure . Despite a greater risk of acquiring leprosy in household contacts , research has found that up to 75% of new cases in a high prevalence area had no known index case [8] . Recent recommendations have been for leprosy control to be focused on high risk contacts as leprosy has become rarer , with the exception of some smaller areas of high prevalence [9] . This has significant implications on whether a focused household or mass chemoprophylaxis would be beneficial in the population . SIMCOLEP is a leprosy microsimulation model which simulates individual life and disease histories [10] . The use of a mathematical models allows for the impact of interventions to be predicted in a specific population over a long time frame , based on past and current epidemiological data . This is important for leprosy as the long incubation period means that it can take many years for the effect of interventions to be visible . Previous applications have been the modelling of household chemoprophylaxis in Bangladesh [11] and Brazil [12] , and the prediction of country-level elimination in India , Brazil and Indonesia . The Kiribati MHMS wished to determine whether chemoprophylaxis could contribute to the control and potential elimination of leprosy in Kiribati . We adapted the SIMCOLEP microsimulation to simulate the demographic characteristics and leprosy control program in Kiribati over time , fitting it to the leprosy new case trend from 1989 . It was then used to compare the predicted effectiveness of household contact , mass , and combined household and mass chemoprophylaxis strategies in reducing future leprosy cases .
The study protocol was reviewed by the senior clinical management team of the MHMS in Kiribati and University of Otago Human Ethics committee where it was regarded as minimal risk health research not requiring informed consent ( June 2018 ) . SIMCOLEP is a stochastic individual-based model that simulates a closed population of fictitious individuals structured into households [10] . These individuals each have a life history containing birth , death , marriage , and children . Households form and dissolve during the simulation with the movement of individuals or couples . The life histories modelled reflect , as far as possible , the realities in the country which is being modelled . The transmission of Mycobacterium leprae is modelled through two processes; general population and within-household . Transmission occurs by direct contact with an infectious person , the rate of which is different in the general population and within-households . The contact rate and the probability of infection during contact determines infectivity . There are two distinct forms of leprosy in this model: PB , which self-heals after some time and multibacillary ( MB ) , the latter being the only infectious form in this model . The natural history of infection is the same as in a previous leprosy model SIMLEP [13] . Susceptibility of individuals was randomly assigned at birth and we assumed that 20% of individuals are susceptible based on the results of model fitting with SIMCOLEP in a previous study [10] . The type of leprosy is randomly determined and in keeping with the Kiribati pattern we assigned two-thirds of cases to be PB . Adapting SIMCOLEP to the situation in South Tarawa was carried out by using available data on the demography of Kiribati , the epidemiology of leprosy there , the history of leprosy control measures , and the calibration of parameters whose values are unmeasured or unmeasurable . Through calibration we derived optimal values for the household parameters , and a range of values for epidemiological parameters . For our predictions , we used the optimal values for household parameters and sampled uniformly from the derived range of values for the epidemiological parameters . This allows our predictions to account for the uncertainty in epidemiological parameter values . First we fitted the model to the years 1989 to 2000 , and omitted the remaining years . Short-term prediction for the years 2001 to 2003 were compared to data to validate the model . The model was deemed valid when it was able to predict the number of cases from 2001 to 2003 with sufficient accuracy to lie within the confidence interval for the prediction . We did not include the years after 2004 because these trend is confounded by substantial changes in the leprosy control situation . Then we fitted the model to all the data , and forward predictions were based on the final fit .
A good model fit to the observed leprosy trend was achieved through calibration ( Fig 1 ) . The model predicts a substantial decrease in cases by continuing the baseline control program . From 2017 to 2030 the number of new leprosy cases will decrease by 88 . 2% ( 95% CI: 85 . 0–91 . 4 ) . The greatest drop is from 2016 to 2017 , with a predicted decrease of 58 . 4% ( 95% CI: 55 . 0–60 . 5 ) . All intervention scenarios are predicted to lead to an even greater reduction from first introduction and over the entire prediction period ( Fig 2 ) , demonstrating the benefit of chemoprophylaxis even when there is a downward leprosy trend . The ranking of interventions by their predicted additional reduction relative to the baseline control is the same for simulations at 100% and 80% coverage of individuals ( S2 Fig ) . As expected , a lower coverage level reduces the impact of the interventions . The following comparisons of interventions are at 80% coverage because it is more realistic . Chemoprophylaxis in household contacts only leads to a slower initial reduction in cases than intervention scenarios with mass chemoprophylaxis only . The predicted number of cases under the household only strategy continues to decrease relative to the baseline control program over most of the prediction period whilst mass only strategies slow down or plateau . One round of mass chemoprophylaxis is the most ineffective intervention scenario simulated because the reduction in cases plateaus from around the year 2020 . Three rounds of mass chemoprophylaxis is predicted to lead to a much greater reduction although the benefit of this also plateaus relative to strategies that include household contact chemoprophylaxis . When interventions are compared ten years after the introduction of interventions , the ranking of interventions by increasing additional reductions relative to the baseline control program is; 1 ) one round of mass chemoprophylaxis; 2 ) household contact chemoprophylaxis only; 3 ) three rounds of mass chemoprophylaxis; 4 ) household contact chemoprophylaxis with one round of mass; 5 ) three rounds of mass chemoprophylaxis with household contact . The combined interventions are more effective at preventing cases when compared ten years after introduction because of the fast initial reduction with mass chemoprophylaxis , and the ongoing nature of household contact chemoprophylaxis . Household contact chemoprophylaxis and one round of mass is predicted to lead to an additional reduction of 77 . 8% compared to the baseline control program , and this increases to 88 . 3% when three rounds are implemented . Comparisons of interventions by the cumulative reduction in cases for each intervention compared to the baseline control program demonstrate that there is little difference between an intensive three rounds of mass chemoprophylaxis ( 49 . 3% ) and one round with household contact ( 47 . 6% ) ( Fig 3 ) . Household contact chemoprophylaxis alone is predicted to result in a reduction of 37 . 7% . The combined strategy of three rounds with household contact is predicted to lead to a 57 . 1% reduction in cumulative cases , representing the avoidance of a significant number of cases compared to the baseline control program and other chemoprophylaxis strategies . Although at 10 years the predicted number of cases with three rounds of mass implemented every second year leads to the same annual number of cases as when implemented in consecutive years , more cumulative cases are predicted to be avoided under the latter ( S3 Fig ) .
In this study , it was possible to closely replicate the leprosy trend from 1989 to 2016 using SIMCOLEP . The number of future new cases in South Tarawa are predicted to decrease substantially under the baseline control program alone , however introducing household contact chemoprophylaxis , mass chemoprophylaxis or a combination of these interventions , are predicted to lead to even more rapid reductions in new cases . The predicted reduction under the baseline control program alone highlights the benefits of the current situation of high levels of awareness of leprosy and its symptoms and therefore the importance of early case detection and treatment with the MDT recommended by WHO [20] . This high level of awareness , case detection and treatment is a prerequisite for all intervention scenarios modelled , and future cases would undoubtedly be greater if this were not to be sustained . Mass chemoprophylaxis targets the whole population and therefore is predicted to lead to a faster reduction in cases in the first few years after implementation than household contact SDR chemoprophylaxis . Targeting the whole population has the advantage of reducing cases among more distant social and neighbourhood contacts , which make up the greater proportion of new cases when leprosy prevalence is high [8] . This modelling also demonstrated that it was beneficial to implement a more intensive approach with more than one round of mass SDR in consecutive years rather than every second year . Each additional round of rifampicin benefitted those who had been infected since the previous round and therefore were conferred no protection from the previous dose . Implementation in consecutive years prevents greater transmission in the intervening time period . Despite household contact chemoprophylaxis leading to a slower decline , cases were predicted to continue to decline under this scenario whilst under mass strategies they plateau or even slightly increase . The benefit of household contact chemoprophylaxis is later in the prediction period when a greater proportion of new cases are in the same household as another case . A combined strategy is likely to be needed for elimination , to rapidly reduce transmission in the general population through intensive mass chemoprophylaxis thus reducing future cumulative cases , whilst providing a sustained and timely response to household contacts who are at highest risk . Predictions for the effectiveness of SDR chemoprophylaxis in household contacts differed in countries where SIMCOLEP modelling has previously been undertaken . Fischer et al . predicted it would lead to a 25% lower NCDR than the baseline control program at 25 years after its introduction in northwest Bangladesh [11] . De Matos et al . predicted it would lead to a 40% reduction over the baseline control program alone after 35 years in Pará State , Brazil [12] . These compare with 77 . 4% after 12 years predicted for South Tarawa in this study . The most likely contributing factor to these differences is the household size distributions of these regions . The average household size in Kiribati is 7 . 2 members compared with 4 . 6 members [21] and 4 . 1 members [22] in northwest Bangladesh and Pará State , respectively . There are a number of limitations related to both the model and the dynamics of the spread of leprosy in Kiribati . The validity of this model relies on the accuracy of the reported number of cases in South Tarawa . The peaks in cases can be mostly explained by changes in active or passive case finding which have been implemented in the past , but not sustained . The first of these included a national screening programme in 1997 which covered the whole of Kiribati , but later case finding activities have been more limited [16] . This has contributed to uncertainty in the calibrated parameter values . For example , the passive detection delay in 1985 was very long so that it is possible that the model’s structure prevented the calibration of a shorter delay . However plausible factors such as stigma , knowledge of health care professionals and accessibility to the health care system could contribute to this delay [23] . The model assumed that only MB leprosy was infectious . Epidemiological studies suggest that PB cases could also be infectious , albeit to a lesser extent [24] . Even if PB is much less infectious than MB disease , PB cases could be a significant contributor to transmission in South Tarawa as two-thirds of cases are this form . Therefore , transmission and the number of cases in the future may be underestimated in this model but this is unlikely to impact on the ranking or trends . An additional complication is the high level of internal migration from the outer islands to South Tarawa which could not be included in the population dynamics of the model . It was therefore not possible to determine the impact of the introduction of more cases into South Tarawa over time . It is possible that net migration into South Tarawa would favour implementing a repeated mass chemoprophylaxis strategy to catch migrants previously exposed in other islands of Kiribati . Finally , we do not know the true mechanisms for heterogeneity of leprosy susceptibility . In this model , susceptibility was assumed to be randomly distributed . SIMCOLEP also allows for the specification of household and/or genetic mechanisms , however a previous study demonstrated that no particular mechanism was the most likely [10] . The random mechanism was found to lead to the fastest reduction and is therefore a best case scenario . This model has provided important insights into the complex transmission dynamics of leprosy in South Tarawa . The absolute number of future new cases is predicted to be very low because although the rate is high , the Kiribati population is small . This means that the difference between predicted impacts is only a few cases between interventions . However , the primary focus of this study was to inform policy by qualitative ranking of interventions relative to the baseline control program , or to each other , rather than the absolute impact of each intervention on case numbers . A recent article by Lockwood et al . expressed concerns regarding SDR chemoprophylaxis , in particular in close contacts [25] . They point out the obvious concern that SDR might promote rifampicin resistance while acknowledging that the actual impact of this is not known . They also mention the ethical dilemma of identifying the disease status of leprosy patients to their contacts . Disclosure of disease status in household contacts has been found to be acceptable in Bangladesh and people from Pacific Island countries living in New Zealand [26] . This concern is addressed with the use of mass chemoprophylaxis as case identification is not necessary . Lockwood et al . also cite the issues of focusing on household contacts when the COLEP study found SDR to be more effective in those with the lower baseline risk . This modelling study addressed this by comparing household contact and whole population approaches , however an underlying assumption of all interventions was that SDR cures 50% of those in the subclinical phase as it was not possible to specify different protective effects by baseline risk . The findings of this study suggest that implementation of SDR chemoprophylaxis to household contacts of new cases , together with at least one round , and preferably more , of mass SDR would give the most rapid reduction in new cases . Associated benefits include reducing burden of disease and limiting the social consequences implicit in a diagnosis of leprosy . Despite increasing calls for a focus of leprosy control of high risk contacts [9] , some populations may still benefit from a whole population approach . The findings of this study could also inform leprosy control policy in similar small but densely populated countries or regions . | Leprosy rates in Kiribati are some of the highest in the world and it is one of the few countries yet to reach the World Health Organization leprosy elimination goal of a prevalence of less than one case per 10 , 000 population . The greatest burden is in the capital South Tarawa and the connected islet of Betio . Interest has increased for the use of chemoprophylaxis , the administration of preventive antibiotics to apparently healthy individuals who may be incubating the disease , which has been demonstrated to be effective in both household contacts and the whole population in two recent studies . In this study we used the individual-based model SIMCOLEP to predict the impact of six difference scenarios using single dose rifampicin ( SDR ) chemoprophylaxis in household contacts and/or the entire population on future new cases in South Tarawa . We found that all chemoprophylaxis strategies were predicted to be more effective than the current control strategy , particularly a combination of household contact chemoprophylaxis alongside three rounds given to the entire population in consecutive years . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"geomorphology",
"medicine",
"and",
"health",
"sciences",
"landforms",
"dose",
"prediction",
"methods",
"topography",
"tropical",
"diseases",
"geographical",
"locations",
"bacterial",
"diseases",
"pharmaceutics",
"kiribati",
"neglected",
"tropical",
"diseases",
"islands",
"bangladesh",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"south",
"america",
"epidemiology",
"brazil",
"people",
"and",
"places",
"asia",
"oceania",
"earth",
"sciences",
"computer",
"software",
"leprosy"
] | 2019 | Predicting the impact of household contact and mass chemoprophylaxis on future new leprosy cases in South Tarawa, Kiribati: A modelling study |
The standard view in biology is that all animals , from bumblebees to human beings , face a trade-off between speed and accuracy as they search for resources and mates , and attempt to avoid predators . For example , the more time a forager spends out of cover gathering information about potential food sources the more likely it is to make accurate decisions about which sources are most rewarding . However , when the cost of time spent out of cover rises ( e . g . in the presence of a predator ) the optimal strategy is for the forager to spend less time gathering information and to accept a corresponding decline in the accuracy of its decisions . We suggest that this familiar picture is missing a crucial dimension: the amount of effort an animal expends on gathering information in each unit of time . This is important because an animal that can respond to changing time costs by modulating its level of effort per-unit-time does not have to accept the same decrease in accuracy that an animal limited to a simple speed-accuracy trade-off must bear in the same situation . Instead , it can direct additional effort towards ( i ) reducing the frequency of perceptual errors in the samples it gathers or ( ii ) increasing the number of samples it gathers per-unit-time . Both of these have the effect of allowing it to gather more accurate information within a given period of time . We use a modified version of a canonical model of decision-making ( the sequential probability ratio test ) to show that this ability to substitute effort for time confers a fitness advantage in the face of changing time costs . We predict that the ability to modulate effort levels will therefore be widespread in nature , and we lay out testable predictions that could be used to detect adaptive modulation of effort levels in laboratory and field studies . Our understanding of decision-making in all species , including our own , will be improved by this more ecologically-complete picture of the three-way tradeoff between time , effort per-unit-time and accuracy .
In order to study this three-way tradeoff , we distinguish between the “baseline” cost of time , and the additional cost of the effort that is invested within each unit of time . The “baseline” cost of each unit of time encompasses all those costs that accumulate at a fixed rate when an individual spends time on a particular activity , regardless of the level of effort per-unit-time that it devotes to that activity . These include , for example ( i ) the costs associated with predation risk ( e . g . in each unit of time spent out of cover assessing potential food sources there is a particular probability that the individual will be spotted and then killed or injured by a predator , which will entail a fitness cost ) and ( ii ) opportunity costs ( e . g . each unit of time allocated to foraging cannot be spent searching for mates ) . The baseline cost of time might increase with the appearance of a predator ( increasing the chance of being predated in each time unit spent out of cover ) , or of a potential mate ( increasing the opportunity cost ) . In contrast , the “cost of effort per-unit-time” is the cost to the animal of the resources under its control which it devotes to a particular task within each unit of time . These are costs over and above the “baseline” cost of time that are incurred as a result of spending additional energetic resources on a higher sampling rate or a lower error rate ( as explained above ) . The cost of effort per-unit-time will therefore increase if a given unit of energetic expenditure becomes more costly in fitness terms , for instance because food is less abundant ( and hence existing reserves are harder to replenish ) or because the individual's energetic reserves are depleted ( making the remaining reserves more valuable ) . Effort per-unit-time and time are thus two different units of investment . Distinguishing between them is important . The baseline cost of time and the cost of effort per-unit-time will both vary depending on the states of both the animal and its environment . This variation is unlikely to be perfectly correlated: for instance , a change in an animal's food reserves does not perfectly predict the risk of predation , and the appearance of a predator does not perfectly predict the value of the animal's reserves . There will therefore be some fluctuation in the relative cost of time and effort . An animal that can modulate the amount of effort it invests within each unit of time can take advantage of this fluctuation in relative cost by substituting between time and effort as one becomes more expensive relative to the other . For instance when time is more expensive the animal can respond by spending less time on the task , but more effort within each unit of time that it does spend . It is thus “freed” from the constraints of the simple speed-accuracy trade off . It does not need to accept the same decline in accuracy that an animal facing that two-way tradeoff would face if it reduced its investment of time to the same degree , because it can increase the effort it expends in each of the remaining units of time , boosting its accuracy . This brings a fitness advantage . Similarly , when effort per-unit-time is more expensive the animal can respond by spending more time on the task , but investing less effort in each unit of that time , which should also bring a fitness advantage . To take a hypothetical example , imagine an animal leaving its nest unguarded to gather a food item from one of two alternative patches . Let us assume that it gains a fitness benefit if it accurately chooses the richest patch . If a nest predator appears nearby , time spent foraging will become more expensive as the nest is more likely to be discovered and depredated in the parent's absence . The optimal strategy for an individual limited to a simple speed-accuracy trade-off will be to spend less time gathering information about the two patches and to return to the nest more quickly , accepting the reduction in accuracy that this will entail but improving the chance of preventing nest predation . However , the appearance of the predator does not affect the marginal cost of effort , for instance the energetic cost of faster neural processing to reduce perceptual errors , or of faster movement between the patches to gather samples more quickly . Both of these factors could increase the amount of information the animal gathers about the patches in a given period of time . The optimal strategy for an animal that can modulate effort in this way will be to spend less time in assessing the two patches , but also to expend more effort in each of those units of time . It will thereby maintain better accuracy than an individual limited to a simple speed-accuracy trade off , and will therefore gain the fitness benefit of foraging from the richest patch more often . In this paper we adapt a canonical model of statistical decision-making , the sequential probability ratio test ( SPRT ) [12] , [13] , to demonstrate that the ability to modulate effort levels does indeed confer a fitness advantage , and therefore that we should expect this ability to have evolved in nature . In the model , we examine effort of the second kind defined above: the investment an animal can make to reduce the perceptual errors that pollute the information it gathers before making a decision . Since both kinds of effort have the same effect of increasing the rate at which the individual gathers information , this choice does not affect the conclusions we can draw from our results and they apply equally to effort of the first kind ( sampling rate ) . Individuals in our model are tasked with making a binary choice about some ( initially unknown ) state of their environment , for instance whether a foraging patch is fruitful or not . They gather noisy evidence from their environment , and use the sequential probability ratio test ( SPRT ) [14] to make their decision . The decision threshold an individual uses in the SPRT is defined by the value of the parameter ( increasing results in an increased investment of time spent gathering information , other things being equal ) . Its effort level is controlled by the parameter ( increasing results in an increased investment of effort per-unit-time ) . See “Methods” for a fuller explanation of the operation of these . A unit of time spent gathering evidence imposes a fitness cost of , where is the baseline cost per-unit-time of time spent gathering information and is the additional cost of effort per-unit-time . Making a correct decision brings a fitness benefit of ( measured in fitness units ) . We define the fitness of an individual as the benefit it obtains across all the decisions it makes , less the total cost it incurs in making those decisions ( see equation ( 8 ) in “Methods” ) .
We calculated the fitness of individuals ( equation ( 8 ) ) across a range of and values , and examined how the optimum values moved as we changed Ct , the baseline cost of time . All the fitness landscapes we examined had a single maximum point ( e . g . Fig . 1 ) suggesting that there is a single optimum decision threshold , , and effort level , , for each set of environmental parameter values . As we increased the cost of time , the optimal value of the decision parameter , ( which controls the decision time , other things being equal ) decreased , while the optimal effort level , , increased ( Fig . 1a–c ) . This is because when the baseline cost of time is high , it pays to spend less time gathering evidence , but more effort ensuring that evidence that you do gather is error-free . Using these fitness landscapes , we then defined two groups of individuals . Type 1 could adapt both their threshold parameter and their effort level to the environmentally-determined optima , while Type 2 had a fixed effort level , with their value of set to the environmentally-determined optimum for . We computed the optimal values of the free decision parameters for different values of , and compared the proportion of decisions that the two types made correctly ( ) , their mean decision time ( ) and their fitness ( ) . Individuals of Type 1 made more correct decisions than those of Type 2 at both and ( Fig . 2a ) . They also increased their decision time more dramatically when the cost of time was decreased , and reduced it more when the cost was increased ( Fig . 2b ) . They also had higher fitness ( Fig . 3 ) . They achieved this fitness gain because they substituted effort for time ( or vice versa ) as their relative costs changed . The central result that we seek to illustrate is shown in Fig . 4: as the baseline cost of time increases , individuals not only reduce their mean decision time , but also increase their mean expenditure on effort per-unit-time , . This option is not available to individuals with a fixed effort level , who cannot make the substitution between effort and time .
Our results provide a specific example that illustrates the broader conceptual point made in the introduction: an optimal decision-maker that can modulate the effort it invests per-unit-time will substitute effort for time when the relative cost of time increases , and this will give it a fitness advantage over individuals that cannot do so . We have shown that this effect is produced in the dynamics of the SPRT when the relative costs of time and effort are varied and individuals are allowed to vary their effort levels . Investments of time and effort per-unit-time both lead to the same benefit: a greater number of accurate statistical samples and therefore a more accurate estimate of the state of the world and a more accurate decision . If the cost of a unit of time varies independently of the cost of a unit of effort , individuals who can substitute one currency for the other are able to exploit whichever one is cheapest . They will substitute effort for time when the baseline costs of time become relatively more expensive , and substitute time for effort when effort becomes more expensive ( Fig . 4 shows this effect in the context of our model ) . In contrast , individuals limited to fixed effort level lack this flexibility . Because they cannot substitute into the cheaper currency , accuracy for them comes at a higher price . They therefore have lower fitness . We hypothesize that control of effort per-unit-time will therefore be common in nature and suggest various avenues for experimental work on this topic ( see Box 1 for our empirical predictions ) . It is notable that equivalent substitutions are well studied in other fields . For instance there is an interesting analogy between the speed-effort-accuracy tradeoff we suggest here and the theory of labor-capital tradeoffs in economics [15] . The role of variable effort in biology may have been overlooked until now simply because the cost of effort per-unit-time varies less in the laboratory than it does in animals' natural environments . We expect natural selection to favor the evolution of mechanisms to modulate effort per-unit-time when four conditions are met . First , we note that there is likely to be a baseline cost of maintaining the ability to modulate effort . This ability will therefore be favored whenever a species' environment varies enough that the benefits of the flexibility that ability brings outweigh this baseline cost . This general problem is analogous to others that are well understood . For instance , insect species with a distinct dispersal morph face a similar question: is environmental variability great enough that the benefits of developing dispersal ability outweigh the fixed costs involved ( let alone the variable ones ) [16] ? We predict that species that occupy niches where there is relevant variation in the cost of time will be more likely to have evolved the ability to modulate their effort levels . Second , the correlation between variation in the marginal baseline cost of a unit of time and variation in the marginal cost of a unit of effort per-unit-time must be less than 1 . Where this is the case , optimally-adapted animals will be able to benefit by responding to rising time costs by substituting into effort and vice versa . In contrast , were the correlation perfect ( equal to 1 ) , the costs of time and effort could not vary relative to one another and so there can be no benefit from switching from one to the other . This is not the same as saying that the total expenditures of time and of effort are uncorrelated . They are obviously tightly correlated . This condition concerns variation in the costs per unit of time , not the total costs summed over time . At present there appears to have been no explicit empirical investigation of the relationship between variation in the marginal cost of time and variation in the marginal cost of effort per-unit-time ( Table 1 gives examples of these costs and possible sources of their variation ) . It seems likely that variation in the marginal costs of time will be only weakly correlated with variation in the marginal costs of effort . There is no literature that suggests , for example , that an increase in the probability of being discovered by a predator in any given unit of time predicts the cost of a unit of energetic expenditure . Of course the presence of a predator is likely lead to a greater total expenditure of energy ( e . g . through fleeing or deterrence ) but this is not the same as suggesting that it would result in an increase or decrease in the fitness cost of spending a single unit of energy . In any case , even if there is a non-zero correlation between changes in the cost of time and the cost of effort per-unit-time , there will be a fitness benefit to substituting between time and effort per-unit-time wherever that correlation is less than perfect . Third , there must be diminishing marginal returns to increasing investment in effort per-unit-time [17] . Diminishing marginal returns are ubiquitous in disciplines concerned with individuals' investment decisions such as biology and economics and would obviously be expected here . However it will be useful to test the magnitude of these diminishing returns explicitly in the case of effort per-unit-time , in order to verify that the relationship between cost and error is concave over the behaviorally relevant range and to parameterize our model . Fourth , the fraction of the sampling error that can be eliminated by increasing effort per-unit-time must be sufficiently large . The noise in the samples an animal gathers is due both to the intrinsic variability of the environment and to the error introduced by its perceptual systems . If the first source of error swamps the second , the potential gains from increasing effort per-unit-time may not outweigh the costs , in which case we would not expect animals to vary their effort levels . Again , the relative magnitudes of these sources of noise have not yet been measured , and this is an interesting area for future work . We predict that species found in environments in which the signals they use for decision-making are noisier will be less likely to have evolved the ability to vary their effort levels . This condition would obviously not apply where the animal increases its effort by gathering more samples per unit time , rather than by reducing the errors in those samples . Decision-making is central to the ability of all animals to survive and reproduce . The current view in behavioral ecology and neuroscience is that animals face a two-way tradeoff between speed and accuracy [1] , [2] , [18] . We suggest that effort per-unit-time , previously neglected , should be added as a third dimension in this tradeoff . If animals can adjust not only how long they spend gathering information prior to making a decision but also the effort they invest on gathering that information within each of those units of time , this can lead to an increase in both their accuracy and their fitness in the face of a changing environment . Current empirical evidence of the link between speed and accuracy in a wide range of species is consistent with this updated paradigm [3]–[8] . However , additional experiments comparing wild-type individuals with others that have had the hypothesized mechanisms of effort modulation “knocked out” would help to determine whether animals adaptively modulate their effort levels . Testing the hypothesis that animals have evolved to balance a three-way trade-off among speed , effort per-unit-time and accuracy will deepen our understanding of decision-making in all species , including our own , and may lead to the development of more efficient control algorithms for artificial decision-makers .
In our model , individuals use the sequential probability ratio test to choose between two hypotheses about their environment [12] . We use a standard choice task in which an individual must decide whether some object is in state 1 or state 0 . In order to inform this decision , the individual gathers a string of samples from the object . These samples are themselves either 1 s or 0 s with a probability that depends on the state of the object . The task here is to determine which of two known distributions the evidence is drawn from . This allows us to isolate the speed-effort-accuracy tradeoff from other factors . A more complex decision task that could also be used in this context is the multi-armed bandit problem . There , an individual is faced with K levers each providing a different ( unknown ) distribution of rewards when pulled , and its goal is to maximize the overall payoff . In the bandit problem the speed-effort-accuracy tradeoff is overlain with a second tradeoff between exploration ( trying different arms ) and exploitation ( sticking with the arm believed to give the best payoff ) [19] . In order make this abstract choice task easier to explain , let us imagine it within an imaginary ecological situation: individuals are foraging in a season where trees may be one of two types , good or bad ( i . e . 1 or 0 ) . They can gather evidence regarding the state of a given tree by examining the fruit husks lying beneath it . An individual will benefit if it can correctly decide whether a tree is good or bad before climbing up to the canopy to forage , but there is a cost to spending time and effort examining husks . Of the fruit husks lying below a good tree a proportion are a deep green color , indicating that they are fine , and contained nutritious material . The others are paler and drier , indicating that the fruit inside was rotten . Under a bad tree , a proportion of the husks are from good fruit . Individual animals know the values of and , but not the state of the tree in front of them ( which in our simulations is good 50% of the time and bad 50% of the time , chosen at random ) . Their task is to decide the tree's type . Each husk therefore constitutes a piece of evidence that the animal gathers , , that has value 1 if the husk is and 0 if it is . The individuals then apply the sequential probability ratio test to the evidence . Each piece of evidence is converted into a weight , which is the log-likelihood ratio in favor of the hypothesis that the tree is good ( ) versus bad ( ) . If the husk was observed to be fine , then the weight of evidence is given by ( 1 ) and if it was observed to be rotten , then the weight of evidence is given by ( 2 ) These weights are summed over time to form the individual's decision variable . We measure time in terms of the number of pieces of evidence observed . After pieces of evidence , the decision variable is given by ( 3 ) The individual's decision thresholds are determined by the parameter . This parameter is bounded so that . An individual decides that a tree is good if ( 4 ) and decides that the tree is bad if ( 5 ) The closer is to 1 , the higher the threshold , and ( with other parameters held constant ) the more pieces of evidence the animal will gather before making a decision . Individuals make errors of perception , misjudging the state of a fraction of the husks they observe . The amount of effort they invest in assessing each husk is given by the parameter . In the real world , effort might include the level of visual or olfactory attention used to examine a husk , the energetic cost of picking it up to assess its mass and the risks associated with tasting it . The value of is the probability of correctly observing the state of a piece of evidence . Each sample takes the true value of husk with probability , and with probability the value of is chosen at random from . The value of is bounded at 0 and 1 . Individuals know their own level of error , and therefore use the adjusted parameters and in the SPRT , where and . Acquiring each sample imposes a cost , where is the cost of the time taken to gather a sample and is the additional cost of the effort invested in ensuring that sample is accurate . The value of is determined by the individual's environment ( so in our model we set it exogenously ) . In contrast , is a function of the amount of effort per-unit-time . We assume diminishing returns to increasing investment of effort and use a hyperbolic function for the relationship between and . The probability of making an error-free observation , , is given as a saturating function of : ( 6 ) where is a parameter that determines how quickly the reduction in error saturates with increasing expenditure on effort . This can be rewritten to give an expression for the cost as a function of : ( 7 ) In any given environment , the average cost an animal pays in order to reach a decision is therefore a function of the two evolving decision parameters , and . Between them , these affect the cost of examining each husk , and the total number of husks examined before a decision is reached . Individuals gain a benefit if the decision they make about the state of the tree they are under is correct . The optimal strategies were found approximately through use of a discrete grid of values for and . We simulated the mean decision time , , and the proportion of correct decisions , , using an ensemble of stochastic realizations of the model for each pair of values on this grid . We then calculated absolute fitness , , as ( 8 ) The optimum strategy for a given set of environmental parameters is given by the pair of values of and that led to the greatest fitness . | Efficient decision-making is vital to the lives of all animals , but the underlying principles of how they achieve this are not yet fully understood . Researchers studying decision-making have generally assumed that animals balance a two-way trade-off between speed and accuracy: the more time they spend gathering information , the more accurate their decisions will be , but the greater the cost they have to pay . We suggest that this picture is missing a crucial component: the effort that animals spend on gathering information within each unit of time . This is important because an animal that can change the amount of effort it invests per-unit-time can use this ability to maintain the accuracy of its decisions even when it reduces the amount of time it spends on them , and can therefore gain a fitness advantage . We predict that this ability to change effort levels should therefore be widespread in nature . This updated view of a three-way trade-off between speed , effort per-unit-time and accuracy will help behavioral ecologists , neuroscientists , economists and psychologists to understand decision-making better , and may also lead to the development of more efficient control algorithms for robot decision-makers . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"psychology",
"animal",
"behavior",
"zoology",
"ecology",
"cognition",
"decision",
"making",
"behavior",
"biology",
"and",
"life",
"sciences",
"cognitive",
"science",
"neuroscience",
"behavioral",
"ecology"
] | 2014 | An Extra Dimension to Decision-Making in Animals: The Three-way Trade-off between Speed, Effort per-Unit-Time and Accuracy |
Meta-analysis of multiple genome-wide association studies ( GWAS ) has become an effective approach for detecting single nucleotide polymorphism ( SNP ) associations with complex traits . However , it is difficult to integrate the readily accessible SNP-level summary statistics from a meta-analysis into more powerful multi-marker testing procedures , which generally require individual-level genetic data . We developed a general procedure called Summary based Adaptive Rank Truncated Product ( sARTP ) for conducting gene and pathway meta-analysis that uses only SNP-level summary statistics in combination with genotype correlation estimated from a panel of individual-level genetic data . We demonstrated the validity and power advantage of sARTP through empirical and simulated data . We conducted a comprehensive pathway-based meta-analysis with sARTP on type 2 diabetes ( T2D ) by integrating SNP-level summary statistics from two large studies consisting of 19 , 809 T2D cases and 111 , 181 controls with European ancestry . Among 4 , 713 candidate pathways from which genes in neighborhoods of 170 GWAS established T2D loci were excluded , we detected 43 T2D globally significant pathways ( with Bonferroni corrected p-values < 0 . 05 ) , which included the insulin signaling pathway and T2D pathway defined by KEGG , as well as the pathways defined according to specific gene expression patterns on pancreatic adenocarcinoma , hepatocellular carcinoma , and bladder carcinoma . Using summary data from 8 eastern Asian T2D GWAS with 6 , 952 cases and 11 , 865 controls , we showed 7 out of the 43 pathways identified in European populations remained to be significant in eastern Asians at the false discovery rate of 0 . 1 . We created an R package and a web-based tool for sARTP with the capability to analyze pathways with thousands of genes and tens of thousands of SNPs .
Genome-wide association study ( GWAS ) has become a very effective way to identify common genetic variants underlying various complex traits [1] . The most commonly used approach to analyze GWAS data is the single-locus test , which evaluates one single nucleotide polymorphism ( SNP ) at a time . Despite the enormous success of the single-locus analysis in GWAS , proportions of genetic heritability explained by already identified variants for most complex traits still remain small [2] . It is increasingly recognized that the multi-locus test , such as gene-based analysis and pathway ( or gene-set ) analysis , can be potentially more powerful than the single-locus analysis , and shed new light on the genetic architecture of complex traits [3 , 4] . The pathway analysis jointly tests the association between an outcome and SNPs within a set of genes compiled in a pathway according to existing biological knowledge [4] . Although the marginal effect of a single SNP might be too weak to be detectable by the single-locus test , accumulated association evidence from all signal-bearing SNPs within a pathway could be strong enough to be picked up by the pathway analysis if this pathway is enriched with outcome-associated SNPs . Various pathway analysis procedures have been proposed in the literature , with the assumption that researchers could have full access to individual-level genotype data [5–9] . In practice , pathway analysis usually utilizes data from a single resource with limited sample size , as it can be challenging to obtain and manage individual-level GWAS data from multiple resources . As a result , pathway analysis often fails to identify new findings beyond what have already been discovered by the single-locus tests . To maximize the chance of discovering novel outcome-associated variants by increasing sample size , a number of consortia have been formed to conduct single-locus meta-analysis on data across multiple GWAS [10–14] . The single-locus meta-analysis aggregates easily accessible SNP-level summary statistics from multiple studies . Similarly , the pathway-based meta-analysis [15–21] that integrates the same type of summary data across participating studies could provide us a greater opportunity for detecting novel pathway associations . Future association studies focusing on identified pathways would have a much-reduced multiple-comparison burden in searching for novel variants with main or complicated nonlinear joint effects on the outcome of interest . In this paper , we developed a pathway-based meta-analysis procedure by extending the adaptive rank truncated product ( ARTP ) pathway analysis procedure [9] , which was originally developed for analyzing individual-level genotype data . The new procedure , called Summary based ARTP ( sARTP ) , accepts input from SNP-level summary statistics , with their correlations estimated from a panel of reference samples with individual-level genotype data , such as the ones from the 1000 Genomes Project [22 , 23] . This idea was initially used in conducting gene-based meta-analysis [24 , 25] or conditional test [26] . As will be shown in the Results Section , sARTP usually has a power advantage over its competitors . In addition , sARTP is specifically designed for conducting pathway-based meta-analysis using SNP-level summary statistics from multiple studies . In real applications ( e . g . , the type 2 diabetes example described below ) , it is very common that different studies could have genotypes measured or imputed on different sets of SNPs . As a result , the sample size used in the pathway-based meta-analysis on each SNP can be quite different . Ignoring the difference in sample sizes across SNPs in a pathway-based meta-analysis would generate biased testing results . Pathway analysis generally targets two types of null hypotheses [4] , including the competitive null hypothesis [15 , 16 , 18–20] , i . e . , the genes in a pathway of interest are no more associated with the outcome than any other genes outside this pathway , and the self-contained null hypothesis [17 , 21] , i . e . , none of the genes in a pathway of interest is associated with the outcome . The sARTP procedure focuses on the self-contained null hypothesis , as our main goal is to identify outcome-associated genes or loci . Also , as pointed out by [27] , tests for the competitive null hypothesis often assume that genotype measured at different genes are independent when evaluating the association significance level . This assumption , which is generally invalid in practice , is unnecessary for sARTP when testing the self-contained null hypothesis . One may refer to [27] and [4] for more discussion and comparison of these two types of hypotheses . The pathways defined in many public databases can consist of thousands of genes and tens of thousands of SNPs . To make the procedure applicable to large pathways , or pathways with high statistical significance , we implement sARTP with efficient and parallelizable algorithms , and adopt the direct simulation approach ( DSA ) [28] to evaluate the significance of the pathway association . We demonstrated the validity and power advantage of sARTP through simulated and empirical data . We applied sARTP to conduct a pathway-based meta-analysis on the association between type 2 diabetes ( T2D ) and 4 , 713 candidate pathways defined in the Molecular Signatures Database ( MSigDB ) v5 . 0 . The analysis used SNP-level summary statistics from two sources with European ancestry . One is generated from the Diabetes Genetics Replication and Meta-analysis ( DIAGRAM ) consortium [13] , which consists of 12 , 171 T2D cases and 56 , 862 controls across 12 GWAS . The other one is based on a T2D GWAS with 7 , 638 T2D cases and 54 , 319 controls that were extracted from the Genetic Epidemiology Research on Aging ( GERA ) study [29 , 30] . The novel T2D-associated pathways detected in the European population were further examined in Asians using summary data generated by the Asian Genetic Epidemiology Network ( AGEN ) consortium meta-analysis , which combined 8 GWAS of T2D with a total of 6 , 952 and 11 , 865 controls from eastern Asian populations [10] .
Here we describe the proposed method sARTP for assessing the association between a dichotomous outcome and a pre-defined pathway consisting of J genes . The same procedure can be applied to study a quantitative outcome with minor modifications .
Firstly , we conducted a simulation study to evaluate the empirical size of sARTP and MsARTP . Secondly , we compared empirical powers of different strategies for carrying out pathway-based meta-analysis that integrated summary statistics from multiple studies . We also evaluated whether results from sARTP were consistent with the ones from MsARTP . Thirdly , we compared our method to the recently developed method aSPUsPath [8] that can be used for pathway-based meta-analysis . We used the R package , aSPU ( version 1 . 39 ) , with the default settings given in [8 , 17] to conduct the aSPUsPath test . To demonstrate the consistency between results obtained by sARTP using SNP-level summary statistics and the ones by ARTP using individual-level genotype data , we compared pathway analysis results from three different procedures on the 4 , 713 candidate pathways using the GERA GWAS data . Details on how those 4 , 713 pathways were pre-processed are given in the Results of T2D Pathway Analysis Section . We applied sARTP to the SNP-level summary statistics generated from the GERA study , using either an internal or an external reference panel . We also obtained the pathway p-values by directly applying the ARTP method to the individual-level GERA GWAS data . Fig 1 shows the comparison among p-values from these three analyses , and demonstrates that all three approaches can generate very consistent results .
The URLs for data and software presented herein are as follows: DIAbetes Genetics Replication And Meta-analysis ( DIAGRAMv3 ) , http://diagram-consortium . org/ Genetic Epidemiology Research on Aging ( GERA , dbGaP Study Accession: phs000674 . v1 . p1 ) , http://www . ncbi . nlm . nih . gov/projects/gap/cgi-bin/study . cgi ? study_id=phs000674 . v1 . p1 Molecular Signatures Database ( C2: curated gene sets ) , http://software . broadinstitute . org/gsea/msigdb/collections . jsp#C2 BioMart ( Homo sapiens genes NCBI36 and GRCh37 . p13 ) , http://feb2014 . archive . ensembl . org/ IMPUTE2 , https://mathgen . stats . ox . ac . uk/impute/impute_v2 . html GWAS Catalog , http://www . ebi . ac . uk/gwas/ 1000 Genomes Project ( Phase 3 , v5 , 2013/05/02 ) , ftp://ftp . 1000genomes . ebi . ac . uk/vol1/ftp/release/20130502/ aSPU , https://cran . r-project . org/web/packages/aSPU/index . html GTEx Portal v6 , http://gtexportal . org/home/ GeneCards Human Gene Database , http://www . genecards . org/ Ingenuity Pathway Analysis , http://www . ingenuity . com/ LocusZoom , http://locuszoom . sph . umich . edu/locuszoom/ ARTP2 package , https://cran . r-project . org/web/packages/ARTP2/ Web-based tool of ARTP2 , http://analysistools . nci . nih . gov/pathway/ | As GWAS continue to grow in sample size , it is evident that these studies need to be utilized more effectively for detecting individual susceptibility variants , and more importantly , to provide insight into global genetic architecture of complex traits . Towards this goal , identifying association with respect to a collection of variants in biological pathways can be particularly insightful for understanding how networks of genes might be affecting pathophysiology of diseases . Here we present a new pathway analysis procedure that can be conducted using summary-level association statistics , which have become the main vehicle for performing meta-analysis of individual genetic variants across studies in large consortia . Through simulation studies we showed the proposed method was more powerful than the existing state-of-art method . We carried out a comprehensive pathway analysis of 4 , 713 candidate pathways on their association with T2D using two large studies with European ancestry and identified 43 T2D-associated pathways . Further examinations of those 43 pathways in 8 Asian studies showed that some pathways were trans-ethnically associated with T2D . This analysis clearly highlights novel T2D-associated pathways beyond what has been known from single-variant association analysis reported from largest GWAS to date . We also identify a novel locus for T2D in the European populations at chromosome 17q21 ( rs1058018 , p = 3 . 06 × 10−8 ) . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"variant",
"genotypes",
"carcinomas",
"cancers",
"and",
"neoplasms",
"gastrointestinal",
"tumors",
"liver",
"diseases",
"genetic",
"mapping",
"epidemiological",
"methods",
"and",
"statistics",
"oncology",
"research",
"design",
"mathematics",
"statistics",
"(mathematics)",
"test",
"statistics",
"genome",
"analysis",
"gastroenterology",
"and",
"hepatology",
"research",
"and",
"analysis",
"methods",
"case-control",
"studies",
"mathematical",
"and",
"statistical",
"techniques",
"epidemiology",
"statistical",
"methods",
"hepatocellular",
"carcinoma",
"epidemiological",
"statistics",
"heredity",
"meta-analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"genomics",
"statistics",
"computational",
"biology",
"human",
"genetics"
] | 2016 | A Powerful Procedure for Pathway-Based Meta-analysis Using Summary Statistics Identifies 43 Pathways Associated with Type II Diabetes in European Populations |
Defining master transcription factors governing somatic and cancer stem cell identity is an important goal . Here we show that the Oct4 paralog Oct1 , a transcription factor implicated in stress responses , metabolic control , and poised transcription states , regulates normal and pathologic stem cell function . Oct1HI cells in the colon and small intestine co-express known stem cell markers . In primary malignant tissue , high Oct1 protein but not mRNA levels strongly correlate with the frequency of CD24LOCD44HI cancer-initiating cells . Reducing Oct1 expression via RNAi reduces the proportion of ALDHHI and dye effluxHI cells , and increasing Oct1 increases the proportion of ALDHHI cells . Normal ALDHHI cells harbor elevated Oct1 protein but not mRNA levels . Functionally , we show that Oct1 promotes tumor engraftment frequency and promotes hematopoietic stem cell engraftment potential in competitive and serial transplants . In addition to previously described Oct1 transcriptional targets , we identify four Oct1 targets associated with the stem cell phenotype . Cumulatively , the data indicate that Oct1 regulates normal and cancer stem cell function .
Mammalian somatic stem cells have been identified in blood , lung , intestine , breast , epidermis and other tissues [1]–[6] . Cancer stem cells ( or cancer-initiating cells ) have been defined in a variety of developmentally heterogeneous neoplasms [7]–[11] . Two functional properties are consistently used to define both normal stem cells and cancer stem cells ( CSCs ) : the ability to self-renew and the ability to generate progeny cells with more differentiated phenotypes [12] . Because CSCs may have metastatic ability , and are thought to be chemo- and radio-resistant and thus provide a reservoir for replenishing tumor mass [13]–[15] , there is interest in identifying cellular activities that regulate CSC populations . Similarly , the central role of somatic stem cells in maintaining tissue homeostasis places priority on identifying cellular activities governing their function . Wnt- , Notch- and Hedgehog-mediated signaling contributes to the maintenance of certain adult somatic and CSC populations [16]–[18] . Identifying additional regulators would allow for robust stem cell identification and provide possible therapeutic targets . The Oct1 transcription factor is widely expressed in adult tissues . It is related to Oct4 , a regulator of embryonic stem ( ES ) cell pluripotency , and has similar in vitro DNA binding specificity [19] . Oct1 enforces poised transcriptional states [20] and promotes a glycolytic metabolic profile associated with dampened mitochondrial function and reactive oxygen species ( ROS ) levels [21] . This metabolic state is emblematic of both tumor cells and stem cells [13] , [22]–[25] . Loss of Oct1 has little impact on cell growth and viability in culture , or on immortalization by serial passage , but antagonizes oncogenic transformation in vitro and tumorigenicity in vivo [21] . Oct1 also promotes resistance to genotoxic and oxidative stresses , a phenotype observed in stem cells [26] . Oct1 message levels are increased in some forms of gastric cancer [27] , however consistent changes in Oct1 gene expression have not been noted for most malignancies . Oct1 protein levels and post-translational modification states in stem cell compartments and malignancy have not been studied . Here we show that Oct1 controls multiple stem cell phenotypes in normal and tumor cells . In epithelial cells , strong Oct1 protein expression spatially correlates with stem cell niches and high levels of ALDH1 , Lrig1 and Lgr5 , known stem cell markers . Elevated Oct1 protein expression also correlates with elevated ALDHHI and CD24LOCD44HI stem cell-like populations in tumor cell lines and primary breast cancer samples , respectively . In contrast , the correlation with Oct1 mRNA expression is poor . Using ALDH and dye efflux activity as readouts , we demonstrate that Oct1 ablation selectively depletes stem-like populations in multiple human tumor cell lines . Abcg2 , Abcb1 , Abcb4 and Aldh1a1 , genes associated with dye efflux and ALDH activity , are direct Oct1 targets . We show that stable Oct1 knockdown in two different tumor cell lines reduces tumor-initiating frequency , while Oct1 ectopic expression increases tumor initiation . Finally , we show that Oct1 deficient fetal liver hematopoietic progenitors manifest engraftment defects in competitive and serial transplantation situations . The results indicate that Oct1 protein can be used as a stem cell marker , and that Oct1 is a normal and malignant stem cell determinant .
We examined Oct1 expression in frozen human colon sections using indirect immunofluorescence ( IF ) . The Oct1 antibody ( Millipore ) did not cross-react to human Oct2 , Oct4 or Oct6 in control experiments ( Figure S1 ) . IF revealed a sub-population of intensely staining Oct1HI cells , one cell removed from the lumen but clearly within the basement membrane that defines the crypt ( Figure 1A , arrows ) . Similar findings were made using immunohistochemistry ( Figure S2 ) . To examine the location and identity of these cells more closely , we used paraffin-embedded normal human colon sections with antibodies to Oct1 and ALDH1 , which has been shown to mark somatic stem cells in multiple compartments , including the colon [9] , [28] , [29] . Oct1 and ALDH1 staining were widely detectable at low levels relative to controls lacking primary antibodies ( not shown ) , consistent with the wide expression of these proteins . However , in a subset of cells , including within gut crypts , more intense co-staining was evident ( Figure 1B ) . Merging the fields confirmed that cells at the crypt base stained strongly for both proteins ( arrows ) . A few cells ( not in the crypt base ) displayed strong ALDH1 expression only ( Figure 1B , asterisk ) . We identified multiple additional examples of staining with only one antibody ( not shown ) , indicating that the results are not due to spectral overlap . We analyzed six independent tissue sections corresponding to 117 crypts , identifying 77% of cells with both Oct1HI and ALDH1HI expression . 19% of cells stained strongly with ALDH1 only and 4% with Oct1 only . These findings identified a high concordance between cells with high Oct1 and ALDH1 protein levels . To extend these findings we also stained frozen mouse colon tissue sections with antibodies against Lrig1 , which is highly expressed in stem cells at the crypt base [30] , [31] . We identified general co-localization using anti-Oct1 and anti-Lrig1 antibodies , though the Oct1 staining was somewhat more restricted compared to Lrig1 ( Figure 1C , arrows ) . Analysis of Lrig1-positive crypts indicated that 19/30 co-stained with high Oct1 expression . No crypts stained with Oct1 only but not Lrig1 . The small intestine crypt is one of the best-characterized stem cell niches . We examined Oct1 expression in formalin-fixed paraffin-embedded mouse small intestinal ( duodenum ) tissue . In this case we used different anti-Oct1 antibodies ( from Bethyl ) . As with colon , IF revealed a sub-population of intense-staining Oct1HI cells at the crypt base ( Figure 1D , arrows ) . To examine these cells more closely , we performed co-localization studies using leucine-rich repeat-containing G protein-coupled receptor-5 ( Lgr5 ) . Lgr5 marks stem cells in the crypt [32] , where it helps transduce Wnt signals [33] , [34] . We used frozen normal mouse small intestine ( duodenum ) sections together with anti-Oct1 and anti-Lgr5 antibodies in IF . Oct1/Lgr5 co-staining was observed ( Figure 1E and F , arrows ) . The signal was not due to nonspecific secondary antibody binding or autofluorescence , as removing either primary antibody eliminated signal only in the appropriate channel ( Figure S3 ) . Analysis of Oct1/Lgr5 co-localization using tissue from a green fluorescent protein-Lgr5 knock-in mouse [32] ( a gift of H . Clevers ) indicated that 27/30 Lgr5-positive crypts co-localized with Oct1 ( 90% ) . For Figure 1 , longer exposure in the Oct1 channel and comparison to samples processed without the Oct1 antibody indicated that Oct1 was expressed in all cells , but much more strongly expressed in stem cells . Another protein , B lymphoma Moloney murine leukemia virus insertion region homolog-1 ( Bmi1 ) , is known to mark a different group of stem cells at the “+4” position in the small intestinal crypt [35] . Using duodenum from a tamoxifen-injected Bmi1-Cre-ER;Rosa26-Cre reporter mouse ( a gift of M . Capecchi ) , we did not observe significant co-localization ( 0/27 Bmi1-positive crypts , data not shown ) . Cumulatively these data indicate that high Oct1 protein levels mark a specific population of normal stem cells in both colon and small intestine . The high Oct1 signal in stem cell compartments was not due to peculiarities with one particular antibody , autofluorescence or spectral overlap . We performed Oct1/ALDH1 IF on malignant human breast carcinoma sections ( estrogen receptorPOS , progesterone receptorPOS , Her2NEG ) . In addition to somatic stem cells , ALDH1 expression marks CSCs , including in breast and colon [9] , [28] , [36] . Dual staining was evident ( Figure 2A , arrows ) though again we observed examples of cells that stained with Oct1 and not ALDH1 and vice-versa ( asterisks ) . These results indicate that Oct1 levels are elevated in a subset of breast cancer cells that also express high levels of ALDH1 . To corroborate these findings , we performed Oct1 Western blotting using primary human metastatic pleural effusion breast carcinoma cells [37] . A variety of subtypes were tested ( see figure legend ) . Unexpectedly , Oct1 protein expression was highly variable . Samples naturally partitioned into Oct1-high and -low categories ( e . g . , Figure 2B ) . We then determined whether Oct1 levels correlated with cancer–initiating cell frequency using CD24/44 as a measure of mammary tumor-initiating cells [6] . Pleural effusions with low Oct1 protein displayed low frequencies of CD24LOCD44HILinNEG cells , whereas those with high Oct1 expression displayed a greater proportion . Examples from each category are shown in Figure 2C . Quantification from 15 samples is shown in Figure 2D . The observed differences were significant ( P = 0 . 059 ) . For the samples shown in Figure 2A , Oct1 mRNA levels were modulated in a similar manner to protein , however for other tested samples , changes in Oct1 message levels were insignificant ( not shown ) . Performing the same analysis comparing CD24/44 with Oct1 mRNA yielded an insignificant P-value ( Figure 2E ) . These results suggest that a combination of transcriptional/RNA regulation , but mostly regulation at the level of protein production or stability , underlies Oct1 variation in human breast cancer tissue . We used human epithelial tumor cell lines in which we could manipulate Oct1 levels by RNAi and retroviral overexpression . High ALDH activity , as measured by Aldefluor , is a marker of CSCs and tumor cell line populations with stem-like properties [9] , [36] . We evaluated ALDH activity following Oct1 RNAi in A549 lung alveolar adenocarcinoma cells infected with pools of lentiviruses expressing scrambled or Oct1-specific shRNAs . Oct1-specific RNAi reduced activity in the main population by less than two-fold as measured by mean fluorescence . However , Oct1-specific RNAi more significantly impacted the number of ALDHHI cells such that the AldefluorHI “tail” collapsed into a more symmetric distribution after Oct1-specific shRNA expression ( Figure 3A ) . Effective RNAi was confirmed by Western blot ( Figure 3B ) . We also studied ALDH activity in two breast cancer cell lines , MDA-MB-231 and MCF-7 , using transiently transfected siRNA pools . Oct1 RNAi again minimally affected activity in the main population while significantly reducing the number of ALDHHI cells ( Figure S4 ) . Therefore , in three tumor cell lines Oct1 ablation reduces ALDH activity most significantly in ALDHHI cells . Because an increase in Oct1 protein levels in stem-like AldefluorHI cells may underlie the selective effects of Oct1 ablation , we used fluorescence-activated cell sorting ( FACS ) to isolate normal A549 cells on the basis of Aldefluor activity and compared endogenous Oct1 protein levels . Oct1 levels were significantly increased in the AldefluorHI population relative to unsorted cells ( Figure 3C ) . In contrast , no difference in Oct1 ( Pou2f1 ) mRNA was observed ( Figure 3D ) . If elevated Oct1 protein is responsible for conferring an AldefluorHI phenotype , elevation of Oct1 protein levels should increase the population of AldefluorHI cells . A549 cells were infected with retroviruses encoding human Oct1 or empty vector controls . Oct1 overexpression did not grossly effect cells as measured by forward/side scatter ( Figure 3E , top panels ) or cell growth or viability [21] , but did increase the proportion of AldefluorHI cells ( Figure 3E , bottom panels ) . We confirmed Oct1 overexpression by Western blot ( Figure 3F ) . These findings indicate that Oct1 controls the setpoint of AldefluorHI vs . AldefluorLO A549 cells . An Oct1 binding site has been identified in the Aldh1a1 immediate promoter region [38] . This site is highly conserved ( Figure 3G ) . We conducted ChIP using normal A549 cells and the Aldh1a1 promoter-proximal region to confirm Oct1 binding . A robust signal was observed using anti-Oct1 antibodies relative to an intergenic region and to an isotype control antibody ( Figure 3H ) . Oct1 has been associated with two transcription cofactors , NuRD ( in a negative regulatory capacity ) and Jmjd1a ( in a positive capacity ) , in different conditions [20] . ChIP using anti-Jmjd1a or anti-NuRD ( Mta2 ) antibodies resulted in strong enrichment of Jmjd1a but not NuRD ( Figure 3H ) , consistent with Oct1 mediating an activation function at Aldh1a1 in A549 cells . To further buttress these findings we studied an independent stem cell marker . Normal and cancer stem cells are frequently dye effluxHI such that incubation with Hoechst results in a fraction of cells ( the side population , SP ) that can be identified by low fluorescence [39]–[41] . Adenosine triphosphate ( ATP ) -binding cassette ( ABC ) multidrug transporters mediate this activity and contribute to the relative resistance to cytotoxic compounds associated with a stem cell phenotype [39] , [42] . A549 cells contain a robust SP enriched in tumor-initiating cells [43] . To determine whether stable Oct1 knockdown selectively alters the SP , we used a previously established A549 inducible shRNA system [21] . A separate A549 clone inducibly expressing scrambled shRNAs was also used . Cells were stained with Hoechst Red , Hoechst Blue and propidium iodide . Dead cells , which were gated out , did not change significantly in the Oct1 depleted condition ( not shown ) . Although the percentage of cells in the SP varied three-fold from experiment to experiment ( and between A549 clones ) , induction of Oct1 shRNA by the addition of doxycycline uniformly and significantly reduced the SP , while minimally affecting the main population ( Figure 4A , bottom panels ) . In contrast , little effect was observed upon doxycycline treatment of cells stably transduced with scrambled shRNAs ( top panels ) . As expected , the SP was also reduced using the efflux transport inhibitor verapamil ( not shown ) . Oct1 knockdown under these conditions was robust ( Figure 4B ) . Averaged data from three independent experiments is shown in Figure 4C . These data show that Oct1 RNAi specifically decreases the SP in A549 cells . ABC transporter G2 ( ABCG2 , also known as Bcrp1 ) regulates dye efflux activity , including in A549 cells [43] , and stem cell chemoresistance [42] . We identified a consensus Oct1 binding element in the human Abcg2 first intron ( Figure 4D ) . ENCODE consortium data [44] indicates that the related Oct4 transcription factor interacts with this region in human ES cells . There is also a perfect octamer in the mouse Abcg2 first intron ( Figure 4D ) . FASTA alignment of the two sequences indicated that the homology is largely limited to the octamer element . ChIP indicated that Oct1 binds the Abcg2 octamer element-containing region in A549 cells ( Figure 4E ) . As with Aldh1a1 , strong enrichment of Jmjd1a but not NuRD was observed ( Figure 4E ) , consistent with Oct1 mediating an activation function at Abcg2 . Previous work identified an oxidative stress response mechanism in which Oct1 phosphorylation alters DNA binding specificity , causing induced Oct1 binding to DNA binding sites more complex than the canonical octamer element [19] . One such site is known as the MORE ( More palindromic Octamer Related Element ) . Using H2O2-treated HeLa cells and ChIPseq , induced Oct1 binding was observed at MORE-containing targets such as Hmgb3 , Blcap , Rras and Rras2 . We identified another MORE sequence in the ABC transporter Abcb1 , at position +250 , and strong Oct1 ChIP enrichment at Abcb1 in HeLa and A549 cells exposed to 1 mM H2O2 ( Figure S5 ) . Another transporter , Abcb4 , is adjacent to Abcb1 on human chromosome 7 . ChIPseq previously identified inducible Oct1 binding to Abcb4 following H2O2 exposure [19] . Inspection of this region revealed a MORE ( not shown ) . We confirmed inducible binding in A549 cells ( Figure S5 ) . The above findings indicate that Oct1 controls multiple markers and activities associated with stem cells , but do not address whether Oct1 controls the stem cell phenotype itself . We previously showed that stable Oct1 RNAi in luciferase-expressing A549 cells reduces tumorigenicity in xenograft assays without affecting growth rates in culture [21] . In these experiments , 2×106 cells were transplanted and tumor mass was partially reduced ( by approximately 60% ) . We hypothesized that differences in tumor initiating frequency underlie this effect . We injected reduced numbers of cells harboring scrambled or Oct1-specific shRNAs into opposite flanks of nude mice . Cells were pre-treated with doxycycline for 48 hr and injected into immunocompromised mice maintained on doxycycline . Using 1×106 cells , both the scrambled and Oct1-specific shRNA-expressing cells engrafted 15/15 recipient mice . A further tenfold reduction resulted in 12/13 mice engrafted using cells expressing scrambled shRNAs while cells expressing Oct1-specific shRNAs engrafted 4/13 mice in the contralateral flank ( Figure 5A , 5B ) . The remaining six mice showed no evidence of engraftment as assessed by visual inspection , palpation or bioluminescence . 50 , 000 cells engrafted poorly ( 4/15 ) using scrambled shRNAs and not at all with Oct1-specific shRNAs . Using even fewer cells , 0/15 mice engrafted regardless of Oct1 status ( Figure 5B ) . These findings allowed us to calculate that Oct1 shRNA reduces the frequency of initiating cells from ∼1/96 , 000 to ∼1/350 , 000 ( Figure 5A ) . We also over-expressed Oct1 in luciferase-expressing A549 cells . In this case no pre-treatment took place and the mice were not administered doxycycline . Oct1 overexpression was moderate ( Figure 3F ) . This level of Oct1 expression leads to a >2-fold increase in TIC frequency ( Figure 5C ) . Similar results were obtained with MDA-MB-231 human breast adenocarcinoma cells and mammary fat pad engraftment . Using Oct1 lentiviral knockdown , TIC frequency shifted downwards from ∼1/50 , 000 to ∼1/175 , 000 ( Figure 5D–5F ) . Oct1 overexpression increased TIC frequency ∼3-fold from ∼1/50 , 000 to ∼1/17 , 000 ( Figure 5G–5H ) . Germline Oct1 deletion results in early embryonic lethality due to defects in extra-embryonic lineages , in particular trophoblast stem cells [45] . A slightly less severe Oct1 deficient allele dies over a wider developmental window beginning at mid-gestation ( E11 . 5 ) and exhibits pale fetal liver , reduced ß-globin gene expression and reduced Ter119-positive cells [46] . Although these phenotypes are consistent with impaired hematopoietic stem cell ( HSC ) function , using this allele it was found that Oct1 deficient fetal livers reconstitute long-term B and T lymphopoiesis in adult recipients [47] and erythropoiesis in lethally irradiated hosts ( not shown ) . To more carefully assess a cell-intrinsic role of Oct1 in hematopoiesis , we performed serial transplants , and competitive transplants using congenic markers . In primary transplants Oct1 deficient fetal liver engrafted sublethally irradiated Rag1−/− primary recipients as evidenced by the presence of B220+ and Thy1 . 2+ Oct1 deficient donor B and T cells in peripheral blood ( Figure 6A , primary transfer ) , consistent with prior findings in which engraftment is stable beyond 16 weeks [47] . Wild-type ( WT ) and Oct1−/− bone marrow from primary recipients also engrafted secondary recipients comparably at 5 weeks ( not shown ) . However at later time points Oct1 deficient cells reproducibly showed nearly complete failure ( Figure 6A , secondary transfer ) . Combined results from 5 independent trials are shown in Figure 6B . Thy1 . 2+ WT or Oct1 deficient fetal liver cells were combined 1∶1 with Thy1 . 1+/Thy1 . 2+ WT bone marrow for competitive reconstitutions . Unlike WT littermate controls , Oct1−/− fetal liver cells engrafted poorly in the presence of WT bone marrow ( Figure 6C ) . Similar results were obtained using the Ly5 marker , which can be used to detect a broader array of blood cells . Ly5 . 1+ C57BL/6 recipient mice were engrafted with WT or Oct1 deficient Ly5 . 2+ fetal liver cells combined 1∶1 with WT Ly5 . 1+/Ly5 . 2+ fetal liver cells . Again , Oct1 deficient cells were found in peripheral blood at 5 weeks , however large defects were observed , specifically using Oct1 deficient fetal liver , at 15 weeks ( Figure 6D ) . Quantified results from four WT and four Oct1 deficient competitive situations is shown in Figure 6E . To test whether the engraftment defect arises from a hematopoietic stem/progenitor cell defect , we analyzed LinNEGSca1POSc-kitPOS ( LSK ) bone marrow hematopoietic precursor cells in the above mice . Defects at least as robust as those seen in peripheral blood were observed in the Oct1 deficient fetal liver cell-derived bone marrow LSK compartment ( two mice in Figure 6F , see Figure 6G for the whole cohort ) . These findings strongly suggest that Oct1 deficient LSK cells are less robust than their normal counterparts in competitive engraftment assays . We also performed primary transplants using freshly isolated fetal liver cells from Oct1 deficient animals or wild-type littermate controls , or the same cells cultured for two days . Freshly isolated Oct1 deficient cells engrafted lethally irradiated recipient animals as before . Culture of the cells in media containing IL-3 , IL-6 and SCF was sufficient to maintain engraftment potential in wild-type cells , but caused complete engraftment failure in Oct1 deficient cells ( Figure S6 ) . Oct1 regulates intracellular redox levels [21] , [26] . Because culture with antioxidants had been found to correct a similar engraftment defect due to ATM deficiency [48] , we incubated the fetal liver cells in cytokine-supplemented media in low-oxygen conditions or in the presence of N-acetylcysteine . Neither treatment restored engraftment potential ( Figure S6 ) .
Here we show that the Oct1 transcription factor ( gene symbol Pou2f1 , not to be confused with the organic cation transporter , Oct1 ) regulates the stem cell phenotype . The expression of Oct1 in multiple tissues , coupled with our findings in both epithelial and hematopoietic cells , indicates that it may control stem cell function in multiple compartments . Epithelial cells in colon and small intestine crypts show variegated Oct1 expression . The observation of variegated Oct1 expression is consistent with work in the developing eye [49] . Cells with high Oct1 protein expression also strongly express known stem cell markers , including Lgr5 . These findings are consistent with a recent study [50] , identifying Oct1 as one of a select group of factors whose protein but not mRNA levels are increased in isolated Lgr5-positive stem cells . We examined four parameters of CSC phenotype and function: CD24/44 levels , dye efflux , ALDH activity and tumor initiating frequency in xenograft models . Elevated Oct1 expression correlates with ALDH1HI cells in tumor sections , and with the contribution of CD24LOCD44HI cells in breast tumor samples . Oct1 loss of function reduces dye efflux , ALDH activity and tumor initiating frequency in tumor cell lines . Oct1 protein levels are elevated in sorted ALDH1HI populations , and forced Oct1 overexpression increases ALDH1HI cells . Although these assays have their individual limitations [e . g . ] , [ 51 , 52] , the common finding of an underlying role for Oct1 suggests that it is a controller of the CSC phenotype . In contrast to Oct1 protein , the association between Oct1 mRNA levels and stem cell phenotypes is poor . This observation is consistent with findings that Oct1 target sites are highly enriched in the promoters of significantly up-regulated genes in lung and breast adenocarcinoma , leukemia , and myeloid leukemia stem cells without concomitant increases in Oct1 mRNA levels [53]–[57] . Elevated expression of Oct1 protein , but not mRNA , in stem cells may be due to increased rates protein synthesis , decreased rates of destruction , or both . Oct1 is known to be ubiquitinated [58] suggesting that regulated protein stability may be important , but the mechanism is unknown . Much of the increased Oct1 protein in stem cells appears to be cytoplasmic ( Figure 1 ) . The role of this cytoplasmic Oct1 is currently unknown , however Oct1 can be regulated at the level of nuclear/cytoplasmic localization [59] , [60] . In addition , transcriptionally active Oct1 residing in the nucleus may be post-translationally modified in a way that alters its activity . Oct1 activity is regulated by cyclic AMP [60] , cellular stress signals [61] and MAP kinase activity [20] , however Oct1 post-translational modification states in stem cells and malignancy have not been carefully studied . Oct1 may control stem cell phenotypes , in part , through its ability to regulate metabolism . Oct1 controls the expression of metabolic genes such as Pcx and Pdk4 and dampens ROS levels [21] , [26] . Reactive oxygen species ( ROS ) negatively modulate stem cell maintenance and self-renewal [62] . Stem cells are frequently characterized by glycolytic metabolic states and low ROS [13] , [24] . Other relevant Oct1 targets include Hmgb3 [61] , with controls HSC function [63] . Here we show that Oct1 also associates with sites in the Abcg2 , Aldh1a1 , Abcb1 and Abcb4 target genes . Loss of Oct1 reduces engraftment potential in competitive and serial hematopoietic repopulation assays , and compromises the LSK stem/progenitor cell compartment in competitive transplants . These results are consistent with fetal HSC deficiency , though because LSK is an impure population it is formally possible that Oct1 deficient HSCs engraft but function poorly . Bmi1 loss of function also results in hematopoietic failure [64] . As with Bmi1 [24] , Oct1 is linked to mitochondrial function and the DNA damage response [21] , [26] , [61] . Hematopoietic defects are less readily apparent with Oct1 as compared to Bmi1 . Both Dnmt1 deficiency and combined FOXO1/3/4 deficiency manifest milder hematopoietic defects similar to Oct1 [65] , [66] . Previous studies identified embryonic pluripotency gene expression signatures in aggressive human breast carcinomas and in myeloid leukemia stem cells without observed Oct4 expression , suggesting a potential role for Oct4 paralogs [55] , [56] . Oct1 and Oct4 share numerous common targets [19] and common modes of upstream regulation [61] . Therefore , an attractive model is that in those phenotypes common to ES cell pluripotency and somatic/cancer stem cells , Oct1 expressed at high levels mediates a subset of Oct4 functions . This may be particularly true if , in response to signals , Oct1 assumes additional or augmented functionalities .
For Figure 1A , frozen sections were fixed using 3 . 7% paraformaldehyde in phosphate buffered saline ( PBS ) , and permeabilized using PBS with 0 . 05% Tween 20 ( Sigma ) . Cells were stained with a mouse anti-Oct1 antibody ( Milipore MAB5434 ) , counter-stained using TO-PRO . Formalin-fixed paraffin-embedded human tissue microarrays ( Imgenex ) were used for images in Figure 1B and Figure 2A . Formalin-fixed paraffin-embedded mouse small intestine tissue blocks were sectioned and used for Figure 1D . Deparaffinization , hydration and antigen retrieval of human formalin-fixed paraffin-embedded tissue sections was performed as follows: Slides were incubated in a dry oven at 62°C for 1 hour , then dewaxed in xylene for 5×4 minutes , and hydrated in 100% , 95% and 75% ethanol for 2×3 minutes each . Slides were immersed in tap water for 5 minutes , then immersed in citrate buffer ( 0 . 01 M , pH 6 . 0 ) , and microwaved on medium power for 5 min , then on low power for 5 min , and immersed in cold PBS . Non-malignant sections were blocked with 10% horse serum for 30 minutes in a humidified chamber , and incubated with rabbit anti-ALDH1a1 ( Abcam ab52492 ) in 1% horse serum for 2 hours at room temperature . Sections were incubated with goat anti-rabbit Alexa488 ( Invitrogen ) for 1 hr at room temperature . Sections were blocked a second time with 1% BSA for 30 minutes at room temperature , then incubated with mouse anti-Oct1 ( Millipore MAB5434 ) in 1% BSA overnight at 4°C . Sections were then incubated with goat anti-mouse Alexa568 ( Invitrogen ) for 1 hr at room temperature . Cells were mounted using media containing DAPI ( Vector ) . For panels 1D and 2A , two mixed rabbit anti-Oct1 antibodies ( Bethyl , A301-716A , A301-717A ) were used . Panel 2A additionally used a mouse anti-ALDH1 antibody ( Becton-Dickinson , BD 611194 ) . Frozen mouse small intestine sections were used in panels 1C , 1E and 1F and were fixed in 1% paraformaldehyde then washed in PBS . Antigen retrieval was performed as published [67] . Mouse anti-Oct1 ( Millipore MAB5434 ) , goat anti-Lrig1 ( R&D Systems AF3688 ) and rabbit anti-Lgr5 ( Abgent AP2745d ) were used with the M . O . M . kit ( Vector labs ) following the vendor protocol . Biotinylated mouse , goat and rabbit secondary antibodies ( 1∶50 dilution ) were added , followed by streptavidin–horseradish peroxidase ( Vector Vectastain Elite ABC kit ) . The signals were enhanced with the TSA kits NEL 741/744 ( Perkin Elmer ) according to the manufacturer's protocol , with fluorescein and Cy3 as the fluorophores . A549 and HeLa cells were maintained in DMEM ( Invitrogen ) supplemented with 10% serum ( 1∶1 calf∶fetal calf , Atlanta Biologicals ) , 6 mM L-glutamine/50 U/ml penicillin/50 µg/ml streptomycin ( Invitrogen ) , and 50 µM ß-mercaptoethanol ( Sigma ) . A549 cells expressing constitutive firefly luciferase and tet-inducible Oct1 shRNA were cultured and maintained as described previously [21] . MDA-MB-231 cells were engineered to express luciferase using a hygromycin-selectable cassette as for A549 cells [21] . Pleural effusions ( PEs ) from breast cancer patients were initially pelleted , and red blood cells were lysed by resuspending the pellet in ACK lysis buffer ( 150 mM NH4Cl , 10 mM KHCO3 , 0 . 1 mM Na2EDTA ) and incubating at room temperature for 10 min . The cells were re-pelleted and washed with DMEM/F-12 medium three times . During these washes , cancer cells were collected with rapid ( 1 minute ) spins to enrich the pellet with tumor organoids . These cells were either immediately frozen in a solution of 10% DMSO/90% FBS , or maintained in DMEM/F12 1∶1 supplemented with 10 mM Hepes , 5% fetal bovine serum , 1 mg/ml bovine serum albumin , 1 µg/ml insulin with transferrin/selenium ( Inivtrogen ) , 0 . 5 µg/ml hydrocortisone and 50 µg/ml gentamycin . All cells were maintained in 5% CO2 and air in a humidified 37°C incubator . Fetal liver cells were genotyped and transplanted , and Rag1−/− recipients analyzed as previously described [47] . Bone marrow cells ( from femur and tibia ) were depleted of red blood cells using ACK lysis buffer for 1 minute at room temperature . T cells were depleted using a biotinylated CD3 antibody ( eBioscience ) and anti-biotin microbeads ( Miltenyi ) . For competitive transplants , 1 . 5×106 Thy1 . 2+ WT or Oct1 deficient fetal liver cells were combined 1∶1 with Thy1 . 1+/Thy1 . 2+ WT bone marrow depleted of CD3+ T cells , and transplanted into Rag1−/− recipients via retro-orbital injection . For serial transplants , 1 . 5×106 bone marrow cells from primary Rag1−/− recipient mice were used in the secondary transplant . Lethal radiation of WT C57BL/6 recipient animals was achieved with a split dose of 2×4 . 5 cGy , spaced 1 hr apart . For CD24/44 staining , PE cells were cultured overnight to allow the epithelial cells to adhere . Cells were then washed with PBS to deplete dead and hematopoietic cells . Epithelial cells were removed with trypsin-EDTA . Cells were stained with 7-AAD , anti-CD24 , anti-CD44 and antibodies against lineage markers ( CD2 , CD3 , CD10 , CD16 , CD18 , CD31 , CD64 , CD140b ) as published [6] . Non-viable and lineage-positive cells were gated out . CD24LOCD44HI cell quantification was determined without prior knowledge of Oct1 expression levels . Aldehyde dehydrogenase activity was measured in cells as described [36] using the Aldefluor kit ( Stem Cell Technologies ) with 125 ng ALDH substrate and 100 mM DEAB ( Sigma-Aldrich ) . Hoechst side population assays were performed as described [68] , with the following modifications: dye incubation was performed in DMEM with 10% FBS , and the buffer for flow cytometry was PBS with 1 mM EDTA/0 . 5 mM EGTA . Cells were cultured in doxycycline ( 2 µg/ml ) four days prior to analysis . Cells were co-stained with propidium iodide and dead cells were gated out from the analysis . Bone marrow LSK precursor cells were identified as described [69] . RNA was isolated using Trizol ( Invitrogen ) , followed by RNAeasy cleanup ( Qiagen ) . cDNA was synthesized using Superscript III and random hexamers ( Invitrogen ) . For Pou2f1 RT-PCR , 100 ng of cDNA was used for quantitative RT-PCR using a LightCycler 480 ( Roche ) . ΔCt values were determined by subtracting input DNA , and ΔΔCt was determined by subtracting the ΔCt value for control primers . The ΔΔCt were converted to fold change using the formula fold change = 2eΔΔCt and were averaged . Sequences for Pou2f1 qRT-PCR were: Pou2f1 forward , 5′ AAAAGAAATCAACCCACCAAGC; Pou2f1 reverse , 3′ GCTAGTCACAAGGCTTGGTGT . Sequences for Gapdh qRT-PCR were: Gapdh forward , 5′ GGCCAAGGTCATCCATGACAA; Gapdh reverse , 3′AGGGGCCATCCACAGTCTTCT . A549 cells were infected with lentiviral particles containing scrambled or pooled Oct1-specific shRNAs ( Santa Cruz ) , and selected using puromycin . Inducible shRNA knockdown of Oct1 using A549 cells transduced with lentiviruses encoding three different shRNAs was described previously [21] . A combination of two rabbit anti-Oct1 antibodies ( Bethyl , A301-716A , A301-717A ) were mixed and used for Oct1 immunoprecipitation . Anti-Jmjd1a and anti-NuRD ( Mta2 ) antibodies were purchased from Abcam . ChIP conditions for the Aldh1a1 and Abcg2 regulatory regions were described previously [20] . Primer sequences for enrichment at Aldh1a1 were: For , 5′ TTGAATCTTCAAATCGGTGAGTAGG; Rev , 5′ AAGTTTAAAGTCAAAGGCTTCCTGC . Primer sequences for enrichment at Abcg2 were: For , 5′ ATGGCTTTACACTTTACCTGATCCC; Rev , 5′ TGAATGACATAGGTAGACCAGCACG . Intergenic primers were from an intergenic region of human chromosome 19 between the Gadd45b and Lmnb2 loci . The sequences were: F2395 , 5′ TTCTATGCCAAGCCCATTCTAGGTC; F2396 5′ GAGAGGCTCTGTCTGAGGTCACG . ChIP grade rabbit control IgG was purchased from Abcam ( ab46540 ) . Luciferase-expressing A549 cells with inducible Oct1 shRNA knockdown [21] were cultured in the presence of doxycycline ( 2 µg/ml ) for 48 hr and the indicated number injected subcutaneously into NCr nude mice ( Taconic ) provided with 2 mg/ml doxycycline in the drinking water two days prior and throughout the assay . Luciferase-expressing MDA-MB-231 cells were infected with Oct1-specific or control lentiviral knockdown constructs and implanted into the 4th inguinal mammary fat pads of nude mice . For both cell lines , tumor engraftment was calculated at 8 weeks . Analyses were computed as previously described [70] . Briefly , analyses used the ‘statmod’ software package for the R computing environment ( http://www . R-project . org ) . Tumor initiating cell frequencies were estimated using a complementary log-log generalized linear model . Two-sided 95% Wald confidence intervals were computed , except in the case of zero outgrowths , when one-sided 95% Clopper-Pearson intervals were used instead . The single-hit assumption was tested as recommended and was not rejected for any dilution series ( P>0 . 05 ) . The study makes use of laboratory mouse models and primary human tissue . The latter were supplied commercially and from institutional samples . In those cases where institutional samples were used , institutional review board approval covering their use is on file at the University of Utah under the authors' names . Similarly , institutional animal care and use committee approval is present for all mouse procedures , and is on file under the authors' names . All procedures conformed to relevant regulatory standards . | Understanding the mechanisms that control stem cell function is a fundamental prerequisite both for the full application of stem cells to regenerative medicine and for a full understanding of the relationship between stem cells and cancer . In this study we show that a transcription factor known as Oct1 is a central regulator of normal and cancer stem cell function . We show that high Oct1 levels are associated with stem cells in multiple normal and malignant settings . Altering Oct1 expression , up or down , correspondingly alters multiple stem cell parameters , as well as stem cell function . We highlight known and identify new target genes Oct1 binds to that are consistent with a role in stem cell function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"cell",
"biology",
"gene",
"expression",
"genetics",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"dna",
"transcription",
"gene",
"function"
] | 2012 | Transcription Factor Oct1 Is a Somatic and Cancer Stem Cell Determinant |
WAGR syndrome is characterized by Wilm’s tumor , aniridia , genitourinary abnormalities and intellectual disabilities . WAGR is caused by a chromosomal deletion that includes the PAX6 , WT1 and PRRG4 genes . PRRG4 is proposed to contribute to the autistic symptoms of WAGR syndrome , but the molecular function of PRRG4 genes remains unknown . The Drosophila commissureless ( comm ) gene encodes a short transmembrane protein characterized by PY motifs , features that are shared by the PRRG4 protein . Comm intercepts the Robo axon guidance receptor in the ER/Golgi and targets Robo for degradation , allowing commissural axons to cross the CNS midline . Expression of human Robo1 in the fly CNS increases midline crossing and this was enhanced by co-expression of PRRG4 , but not CYYR , Shisa or the yeast Rcr genes . In cell culture experiments , PRRG4 could re-localize hRobo1 from the cell surface , suggesting that PRRG4 is a functional homologue of Comm . Comm is required for axon guidance and synapse formation in the fly , so PRRG4 could contribute to the autistic symptoms of WAGR by disturbing either of these processes in the developing human brain .
The Commissureless protein ( Comm ) in Drosophila regulates the cell surface expression of Roundabout ( Robo ) axon guidance receptors by targeting Robos for degradation during secretion through the ER/Golgi network [1] , reviewed in [2] . Failure to down-regulate Robo leads to a dramatic phenotype in which axon crossing of the CNS midline is abolished [3] . Conversely , overexpression of comm induces ectopic midline crossing through increased removal of Robos [4–6] . Comm is also required for the correct formation of the Drosophila brain commissure [7] . Comm is a relatively short protein with a single transmembrane domain and L/PPxY motifs [1 , 8] . Comm binds the WW domain containing ubiquitin ligase Nedd4 via L/PPxY motifs [9] , but this function appears only to be required for endocytosis activities at the neuromuscular junction [10 , 11] . Despite the conservation of the Robo/Slit pathway , homologues of Comm have not been found outside of insects and alternative molecules and mechanisms have been proposed for Robo regulation in the vertebrate spinal cord [12–15] . The vertebrate proline rich and Gla domain genes PRRG1-4 , also known as PRGP1 , PRGP2 , TMG3 and TMG4 respectively [16 , 17] , encode short transmembrane proteins . PRRG4 protein has been found in the Golgi apparatus and at the cell surface [18–20] . All PRRG proteins contain a Gla domain in which glutamic acid ( Glu ) residues are γ-carboxylated in the endoplasmic reticulum by γ-glutamyl carboxylase ( GGCX ) [21 , 22] to form γ-carboxyglutamate ( Gla ) residues . Gla domains coordinate calcium ions to allow binding to membrane phospholipids [23] . Although γ-carboxylation plays a major role in blood clotting , the enzymes required for this post-translational modification are also found in invertebrates , which lack the vertebrate blood clotting cascade , suggesting additional functions [24 , 25] . PRRG proteins are expressed highly in tissues such as the spinal cord and so are believed to play roles outside the coagulation cascade [16 , 17] . The cytoplasmic domains of PRRG proteins are characterized by PPxY and LPxY motifs that are best known as acting as ligands for WW domain containing proteins [26 , 27] . The PRRG proteins are therefore members of a family of transmembrane proteins that can recruit additional proteins or vesicles to the membrane via the Gla domain or L/PPxY motifs . WAGR ( Wilm’s tumor , Aniridia , Genitourinary malformations and mental Retardation ) syndrome is a rare genetic disorder caused by haploinsufficiency of the 11p13 chromosomal region [28–30] . The WAGR critical region includes the WT1 and PAX6 transcription factors , which are responsible for the Wilm’s tumor and aniridia phenotypes respectively [31 , 32] . WAGR syndrome is frequently accompanied by developmental delay and autism like features . The genes that could contribute to these symptoms include PAX6 , SLC1A2 , DCDC1 and PRRG4 [33 , 34] . In a survey of 31 WAGR patients with autism , all were deleted for PRRG4 , a correlation that suggested that PRRG4 is involved in autistic symptoms [33] . The critical region for severe developmental delays and autistic behaviors was subsequently narrowed down to 1 . 6Mb that includes PRRG4 , but not SLC1A2 or DCDC1 [35] . Understanding the function of PRRG4 is therefore a key step in determining whether PRRG4 contributes to the autistic behaviors . During literature searches for short transmembrane proteins containing L/PPxY motifs , we noticed similarities between Comm , the Rcr1 and Rcr2 genes in yeast , and the PRRG , CYYR and Shisa families in vertebrates ( Fig 1A ) . We tested representatives of these families for the ability to affect axon guidance in the fly ventral nerve cord . We find that expression of PRRG4 in a sensitized background induces midline crossing . When expressed in COS cells , PRRG4 reduces the surface localization of Robo proteins . Our results place PRRG4 in an evolutionarily conserved gene family that regulate the cellular localization of cell surface proteins .
The GLPSYDEAL motif of Comm has been shown to be essential for Comm function in midline crossing [1] , and constitutes an extended version of an L/PPxY motif [36] ( Fig 1B ) . We searched the literature for PY motif proteins from other species and compared their structure to that of Comm . In S . cerevisiae , the Rcr1 and Rcr2 proteins contain PPSY and VPEY motifs and have an overall structure resembling that of Comm . The VPEY motif binds the Rsp5 ubiquitin ligase with PPSY having a cooperative function [37] . This activity is likely required for endocytotic trafficking of yeast membrane proteins [38] . In Drosophila , the Nedd4 ubiquitin ligase binds Comm by either the LPSY or PPCY motifs , but with an in vivo preference for LPSY [9] . The Nedd4 interaction is required for endocytosis at the neuromuscular junction formation [11] , but not the regulation of Robo during midline crossing [10] . In an interesting parallel , Rsp5 is not required for activity of Rcr1 in chitin deposition . These similarities led us to test Rcr1 and Rcr2 for activity in the fly nervous system . Yeast has been used to screen for human genes regulating plasma membrane protein trafficking and CYYR1 gene was identified in this manner [39] . CYYR1 is characterized by a cysteine ( Cys ) rich N-terminal , three conserved Cys residues within the transmembrane domain as seen for Comm2 and other insect proteins , and three PPxY motifs ( Fig 1A ) . CYYR1 appears to be a member of the large Shisa-like protein family ( STMC6 ) , all of which are short single pass transmembrane proteins involved in protein trafficking and degradation [40] . Shisa proteins physically interact with Frizzled and FGF receptors in the ER/Golgi , preventing their maturation and trafficking to the cell surface in Xenopus and mice [41 , 42] . Disruption of these developmentally important pathways could potentially mask subsequent effects on axon guidance . However , Comm proteins lack Cys residues in their extracellular domain so are less likely to be homologues . We tested two divergent members , Xenopus Shisa4 and human CYYR1 to check for the ability to regulate Robo . After testing these genes , we observed that the uncharacterized gene CG15760 is likely the Drosophila homologue of Shisa-like gene family , based on the C*C*CC*CC arrangement of Cys amino acids in the putative extracellular/lumenal domain ( S1 Fig ) [40] . Searching for other PY motif proteins , our attention was drawn to the PRRG proteins , two of which lack signal sequences like Comm . All have PPxY and LPxY motifs in their cytoplasmic domains , with PRRG4 having an exceptional match to the critical Comm GLPSYDEAL motif: GLPSYEQAV , when conservative substitutions for the negatively charged and hydrophobic amino acids are taken into account ( Fig 1B ) . The human PRRG2 LPxY sequence closely matches that of Comm homologues from the housefly and the Mediterranean fruit fly . The PPxY motif comes after the LPxY motif in these genes , and an SH3 binding motif is also present in PRRG2 and PRRG4 [17 , 18] . Finally , we noticed an uncharacterized C . elegans gene C17G10 . 7 with two PPxY motifs , one with acidic residues following the tyrosine ( S2 Fig ) . However , an alternative alignment for the predicted C17G10 . 7 protein has four putative transmembrane domains so may align with LAPTM4 proteins instead ( S2 Fig ) [39] . Nevertheless , the predicted protein had additional homologies at the N- and C- termini that led to it being included in testing . As noted in the introduction , PRRG proteins contain an N-terminal Gla domain consisting of Glu residues that are γ-carboxylated by GGCX . The GGCX and VKOR enzymes required for γ-carboxylation are present and functional in flies , but surprisingly GGCX knockouts have no apparent phenotypic defects [43–45] . Gla domains contain a propeptide sequence bound by GGCX , a hydrophobic region called the “keel” or ω-loop that binds phospholipids giving Gla domains membrane binding properties [46] , and a highly conserved region of Glu and Cys residues that coordinate calcium ions . The activity of GGCX on its substrates is greatly enhanced by the presence of a propeptide sequence that is proteolytically removed after GGCX has moved along the protein [47 , 48] . The propeptide consensus consists of a highly conserved phenyalanine residue at -16 , an alanine at -10 and a leucine at position -6 relative to the proteolytic cleavage site , as well as additional conserved hydrophobic amino acids [49 , 50] . An N-terminal motif , ITFEIP , conserved among Comm proteins is centered on a Phe residue and is followed by Ala and Leu residues only slight offset from the vertebrate consensus , suggesting this region could function as a propeptide ( Fig 2B ) . GGCX functions in a processive manner and usually begins modifying Glu residues immediately downstream of the propeptide , which frequently occur within the keel or ω-loop . The sequence FLEEL in PRRG3 represents this initial substrate and is identical to a sequence frequently used to measure GGCX activity and the influence of the propeptide [51] . Comm proteins show distant homology to this initial substrate , although the Glu residues are missing ( Fig 2B ) . The keel region may insert directly into the membrane being bound by the Gla domain , so the hydrophobic residues are likely the most important [52 , 53] . Deletion of this region of Comm greatly reduces Comm function in vivo indicating its importance [54] . The remainder of the Gla domain coordinates calcium ions via the Gla residues . In Comm , a short sequence adjacent to the transmembrane domain is essential to Comm activity ( labeled the “sorting sequence” in Fig 2B ) [10] . As before there is weak homology to the Gla domain ( Fig 2C ) , with the proposed ω-loop and the rest of the domain physically separated in Comm . Given the distant homologies to Gla domains in Comm , as well as the conservation of the LPxY motif , we tested PRRG1-4 genes in the fly nervous system . The open reading frames of the selected genes ( S . cerevisiae Rcr1 , Rcr2 , C . elegans C17G10 . 7 , X . laevis Shisa4 , and Mus musculus CYYR1 and PRRG1-4 ) were synthesized with a Drosophila codon bias . All open reading frames had a myc epitope tag added at the carboxy-terminus , were subcloned into the pUAST expression vector and used to generate transgenic fly lines . The lines were tested by pan-neural expression using the scabrous-GAL4 ( sca-GAL4 ) driver and staining for the myc epitope to confirm expression . Comm protein is found in cell bodies , cytoplasmic vesicles and axons [8] , and we expected that a candidate homologue might show the same pattern . As yeast lacks a nervous system , we did not expect to see axonal localization of Rcr1 or Rcr2 . However , we found that by stage 16 of embryonic development yeast Rcr1 and to a lesser extent , Rcr2 , localized to longitudinal axons in a manner reminiscent of Robo1 protein ( Fig 3B and 3C ) . This raises the possibility that the Rcr proteins may be weakly interacting with Robo proteins . In contrast , the vertebrate PRRG4 protein remained in the neuronal cell bodies ( Fig 3D ) . As trafficking and cell surface localization of Gla domain proteins can be dependent on γ-carboxylation [55 , 56] , it is possible that the fly GGCX enzyme does not properly process PRRG4 . Pan-neuronal over-expression of comm in the fly CNS induces ectopic midline crossing that phenocopies robo mutants because Robo proteins are downregulated by excess Comm [4–6] . In our hands , CNS axon guidance phenotypes require multiple copies of the sca-GAL4 driver and the UAS-comm transgene . We screened several independent UAS transgene insertions for each candidate Comm homologue by crossing to sca-GAL4 , recovering the F1 generation and examining the embryos laid . This allowed us to rapidly generate large numbers of embryos potentially carrying more than one copy of the sca-GAL4 driver and/or the UAS transgene . Staining of the CNS axon scaffold revealed no mis-expression phenotypes for the CYYR1 , xShisa4 , C17G10 . 7 , Rcr1 , PRRG1 , PRRG2 and PRRG3 genes . Rcr2 expression resulted in very minor aberrations in the axon scaffold in a very low percentage of embryos . PRRG4 expression had very rare and subtle phenotypes ( Fig 4B ) , but still stood out from the other transgenes for having a noticeable effect . Very rarely stronger effects ranging from increased midline crossing in single segments to missing commissures were observed . The latter phenotype suggests that PRRG4 might act as a dominant negative . We repeated the PRRG4 experiments with the scratch-GAL4 promoter , which has a similar expression pattern as sca-GAL4 , but may express for longer , but saw no increase in phenotypes . The low frequency of PRRG4 phenotypes suggested that two copies of the GAL4 and UAS transgenes are required to obtain phenotypes . Increasing expression of the PRRG4 transgene beyond two copies would have been challenging , so we sought out alternative approaches to increase the phenotypic penetrance of PRRG4 expression . We were concerned that interactions between Robo and Comm might be species specific , as Comm has no effect on zebrafish Robo1 or Robo3 localization in S2 cells [57] . Expressing human Robo1 ( hRobo1 ) in the ventral nerve cord subtly increases midline crossing ( Fig 4C , Table 1 ) . This is in contrast to fly robo1 over-expression , which leads to a commissureless phenotype [58] . Fly and vertebrate Robo proteins can dimerize via their cytoplasmic and extracellular domains [6 , 59 , 60] , and also form heterodimeric complexes with other receptors bridged by Slit [61] . This suggests that hRobo1 may be acting as a dominant negative , interfering with the function of endogenous Robos perhaps by creating inactive heterodimers . Co-expression of PRRG4 with hRobo1 strongly enhanced the midline crossing phenotype ( Fig 4D , Table 1 , S1 Data ) . The interaction of the γ-carboxylated PRRG4 protein and hRobo1 suggested that γ–carboxylation might be important for fly nervous system formation . We examined the nerve cords of mutants for the γ-glutamyl carboxylase ( GC ) gene , but found no defects in the axon scaffold ( Fig 4E ) . In these embryos co-expressing PRRG4 and hRobo1 , fly Robo1 protein can be found in the commissures ( Fig 4F and 4H ) . A similar mislocalization is seen in comm gain of function embryos [4] . We examined the protein localization of hRobo1 when expressed in the fly ventral nerve cord and found it present in the commissures suggesting it is not regulated by fly Comm ( Fig 4J ) . Of the candidate genes tested , PRRG4 was the strongest candidate for a Robo regulator identified in these tests . To further investigate the potential PRRG4-Robo interaction , we co-expressed constructs in COS cells and looked for co-localization . We began by testing Comm and rat Robo1 ( rRobo1 ) . Robo proteins localize to the cell surface , whereas Comm is primarily in the ER/Golgi ( Fig 5A ) [1 , 8] . Some co-localization occurs but may be because both proteins are in the secretory pathway . The clearance of dRobo1 from the cell surface of COS cells has been used as an assay for Comm function [1 , 62] , but we saw no evidence that rRobo1 is cleared from the cell surface by Comm suggesting that these two proteins do not interact . Similarly , co-expression of PRRG4 and fly Robo1 ( dRobo1 ) showed limited co-localization and no re-localization of dRobo1 from the cell surface ( Fig 5B ) . Taken together with previous results showing no interaction between Comm and zebrafish Robo1 and Robo3 in S2 cells [57] and our results showing no localization of hRobo1 in the fly ventral nerve cord ( Fig 4J ) , this suggests that interactions between Robo and Comm/PRRG genes have co-evolved since insects and vertebrates split . We tested all four mouse PRRG proteins for co-localization with rRobo1 . PRRG1 and PRRG2 showed minimal or no co-localization with rRobo1 , and rRobo1 did not appear to re-localize from the cell surface ( Fig 5C and 5D ) , suggesting these proteins do not interact . We obtained mixed results with PRRG3 as we saw partial co-localization with rRobo1 , but also clear separation of staining ( Fig 5E ) . rRobo1 also appeared to be partially cleared from the cell surface ( Fig 5E” ) , and these results may be interpreted as a weak interaction between PRRG3 and rRobo1 . We have included additional examples of co-localization to document this effect ( S3 Fig ) . PRRG4 showed a strong co-localization with rRobo1 , clearing rRobo1 from the cell surface and co-localizing in the presumed ER/Golgi adjacent to the cell nucleus ( Fig 5F ) . This result strongly resembles that of Comm and dRobo1 , suggesting that PRGR4 and rRobo1 interact in cell culture . To verify the co-localization results , we chose to test the PRRG proteins' ability to clear rRobo1 from the cell surface in a blinded experiment in which COS cells were co-transfected with both genes of interest but the experimenter responsible for scoring only observed the dRobo1/rRobo1 staining . Comm and dRobo1 served as a positive control and Comm re-localized dRobo1 from the cell surface with 100% efficiency when scored blind ( Fig 6A–6C; S2 Data ) . When rRobo1 was co-expressed with the PRRG genes , PRRG4 prevented cell surface localization of rRobo1 or showed increased rRobo1 localization in the ER/Golgi over 80% of the time ( p < 0 . 0001 , two tailed Fisher’s exact test; Fig 6G; S2 Data ) . None of the other PRRG genes had a statistically significant effect on rRobo1 localization , although PRRG3 trended towards statistical significance , ( p = 0 . 0538 , cutoff value is p < 0 . 0125 , Fig 6F ) , consistent with the mixed results obtained in the co-localization assay . A dosage sensitive relationship between Comm and Robo has previously been demonstrated in cell culture , with increasing amounts of Robo plasmid leading to less Comm protein detectable by immunoblot [62] . We modified this assay to verify the PRRG4 result and found that increasing amounts of PRRG4 expression reduced rRobo1 levels as detected by immunoblot ( Fig 7; S3 Data ) . We used the related immunoglobulin family member hDscam as a control and found negligible downregulation in the presence of PRRG4 . 250ng of PRRG4 plasmid per well ( 9 . 5cm2 ) produced a very reliable down-regulation of rRobo1 compared to hDscam ( p = 0 . 00002 , one-way ANOVA , Fisher LSD test ) . Together these results indicate that PRRG4 downregulates Robo in COS-7 cells in a manner analogous to Comm .
The PRRG4 gene has been implicated in the autistic features of WAGR syndrome . Our work suggests that PRRG4 is a functional homologue of the Drosophila commissureless gene and may regulate the cell surface localization of the Robo guidance receptors and other molecules during human brain development . How could haploinsufficiency for PRRG4 lead to autistic symptoms ? The simplest explanation is that reduction in PRRG4 levels alters connectivity patterns in the developing brain due to increased Robo levels . Connectivity defects have been suggested as potentially underlying some cases of autism [63 , 64] . Robos have been implicated in autism through single nucleotide polymorphism and expression studies [65–67] . It has been proposed that Robo gene variants are interfering with the serotonergic system , the anterior cingulate cortex or through a general effect on neurodevelopment . Additionally , alterations to the corpus callosum have also been implicated in autistic symptoms [68] , and Robo/Slit signaling is required for corpus callosum formation [69 , 70] . Robo /Slit signaling has been implicated in all aspects of neural development , not just axon guidance [71] , so it is unclear at what stage of development PRRG4 function might be required . There is little information on the expression pattern of PRRG4 in the embryo , with the exception of Xenopus embryos in which expression appears quite broad and likely to include the CNS [72] . PRRG4 expression has been observed in Purkinje cells in the human cerebellum [19] , neurons known to be important in autism models [73] . Embryonic comm expression is highly dynamic in the fly [1 , 8] , so thorough surveys of PRRG4 expression will be required to identify candidate regions for further analysis . In parallel , the development of knockout mice may also help identify affected brain areas . Identification of a PRRG gene that is expressed in spinal cord commissural neurons during axon crossing of the CNS midline would also establish whether the most well-known function of comm is conserved . Comm is also required for the formation of Drosophila neuromuscular synapses , and is proposed to clear molecules from the cell surface to allow synaptogenesis to take place [11 , 74 , 75] . As autism appears to primarily be a synaptic disorder [76] , haploinsufficiency for PRRG4 may disrupt synapse formation in WAGR syndrome . The synaptic function of Comm in flies has not been linked to regulation of Robo and likely involves unidentified molecules . The ubiquitin ligase Nedd4 is important for the synaptogenesis function and PRRG4 also binds Nedd4 proteins [18] . Additional proteins that interact with the PY motifs of PRRG4 have been observed , including the MAGI proteins , which are required for learning and memory [18] . PRRG4 could function as an adaptor protein regulating molecules acting at the synapse . Our findings suggest that Comm may be γ-carboxylated and that γ-carboxylation could have arisen as a nervous system post-translational modification that was later co-opted for blood clotting . Surprisingly , an absence of γ-carboxylation leads to no phenotypic defects in the fly ( Fig 4E ) [44] , and we have observed no effects of warfarin on embryonic development . If Comm is γ-carboxylated , then this modification is not required for embryonic function . In commissural neurons , Comm sorts Robo into vesicles destined for late endosomes and the lysosome [10] . Sorting may not require γ-carboxylation of Comm , or alternatively the putative Gla domain may have additional functions . We favor a model in which the cell surface localization of Comm/PRRG proteins will require γ-carboxylation and whereas trafficking from trans Golgi network to the lysosome will not . Comm and PRRG proteins have been studied independently up to this point . The existence of molecular and genetic datasets for both genes will aid future experiments into the functions of these protein families . For example , the LPSY motif that binds Nedd4 is also required for Comm function in midline crossing . Additional binding partners for the LPSY motif have been identified [18] , and these can be tested for functions in the fly CNS . Similarly , studies of comm in Drosophila and other species can guide expectations of PRRG4 function in WAGR syndrome [77 , 78] . We were surprised to find that PRRG3 did not interact with rRobo1 at a statistically significant level in the cell clearance assay as we observed partial co-localization ( Fig 5E ) . PRRG3 may be able to regulate Robo proteins in the exocytosis pathway , but less efficiently in endocytosis and will deserve further investigation . Interestingly , PRRG3 and PRRG4 both share a conserved cysteine in the transmembrane domain with insect Comm homologues ( Fig 2C , highlighted in red and blue ) , whereas PRRG1 and PRRG2 do not . Our results suggest that Comm/PRRG proteins are part of an ancient family of cell surface protein regulators that originated in single celled eukaryotes and that a subset of WAGR syndrome symptoms are likely due to increased levels of cell surface proteins in axons or synapses .
The coding sequences of candidate genes Rcr1 ( NM_001178353 ) , Rcr2 ( NM_001180311 ) , C17G10 . 7 ( NM_062689 ) , CYYR1 ( AF442733 ) , xShisa4 ( NM_001096205 ) , PRRG1 ( NM_027322 ) , PRRG2 ( NM_022999 ) , PRRG3 ( BC137616 ) and PRRG4 ( NM_178695 ) were synthesized with codon optimization for expression in Drosophila by Genscript . A C-terminal Myc epitope tag was added to each sequence and genes were delivered in pUC-57 with 5’ and 3’ restriction sites added to facilitate cloning into pUAST . Rcr1 , Rcr2 , CYYR1 , xShisa4 and C17G10 . 7 coding sequences were subcloned into pUAST with EcoRI and XbaI . The PRRG1-4 coding sequences were inserted as EcoRI-KpnI fragments . Drosophila injections were performed by Genetic Services Inc . or Rainbow Transgenics and transformants were selected and insertions mapped using standard methods . For construction of UAS-hRobo1 , the human Robo1 clone described in Kidd et al . 1998 ( Genbank #AF040990 ) was modified by PCR to change the stop codon to leucine ( TGA to TTA ) , thereby introducing a HindIII site at the carboxy terminus of the protein . The original intention to insert an epitope tag appears to have failed . The hRobo1 gene was subcloned into the pUAST vector as an XbaI-HindIII fragment and used to transform Drosophila by standard techniques . scabrous-Gal4 and scratch-Gal4 were obtained from the Bloomington Drosophila Stock Center . Drosophila embryos were processed and immunostained as previously described [79] . The following antibodies were used: mouse anti-c-Myc 9E10 ( Santa Cruz ) 1:200 , BP102 ( DSHB ) 1:10 , mouse monoclonal antibody 13C9 against fly Robo1 ( DSHB ) 1:20 , rabbit anti-hRobo1 ( Abcam ab7279 ) 1:1000 . Anti-mouse ( 1:500 ) and rabbit ( 1:1000 ) HRP- conjugated secondary antibodies were obtained from Jackson Laboratories . For phenotypic comparisons , transgene presence was confirmed by immunostaining . C-terminal myc-tagged Rcr1 , Rcr2 , PRRG1 , PRRG2 , PRRG3 and PRRG4 synthetic sequences were subcloned into pcDNA3 . 1 ( Life Technologies ) for expression in mammalian cells . Drosophila Robo1 in pcDNA is described in [80] . HA-tagged rat Robo1 in pCS2+ was a gift from Yi Rao ( National Institute of Biological Sciences , Beijing University ) to Grant Mastick ( University of Nevada , Reno ) . GFP-Comm in pcDNA was provided by Daniela Rotin ( Peter Gilgan Centre for Research and Learning , Toronto ) . Myc-tagged human Dscam in pcDNA was a gift from K . -L . Guan ( Pharmacology , UCSD ) . COS-7 cells were transfected using Lipofectamine 3000 ( Life Technologies ) and analyzed 48 hours post-transfection for all cell culture experiments . To assay re-localization of Robo in response to Comm/PRRG proteins , 500 ng of Robo plasmid alone or with 250ng candidate gene plasmid were added to each well of a six well plate . For immunocytochemistry , cells were washed with PBS then fixed in 4% PFA . Cells were blocked in 5% NGS for 30 minutes prior to antibody labeling . Antibodies used for immunocytochemistry were mouse anti-c-Myc 9E10 ( Santa Cruz ) 1:200 , rabbit anti-HA ( Covance ) 1:250 , mouse monoclonal antibody 13C9 against fly Robo1 ( DSHB ) 1:20 . Secondary detection used Alexa Fluor anti-rabbit 488 and anti-mouse 568 ( Jackson Laboratories ) . To assay total levels of rRobo1 and hDscam protein in the presence of PRRG4 , 500ng of rRobo1 or hDscam plasmid alone and with increasing amounts of PRRG4 plasmid were transfected per well of six well plates , as described in [62] . After 48 hours cells were harvested and lysed in ice cold lysis buffer containing 50mM HEPES ( pH 7 . 2 ) , 100mM NaCl , 1mM MgCl2 , 1mM CaCl2 and 1% NP-40 with protease inhibitors [81] . Total protein content was normalized using a BCA Protein Assay Kit ( Thermoscientific , Pierce ) . Protein was separated on a 4–20% gradient SDS-PAGE gel and electroblotted to nitrocellulose membrane ( Bio-Rad ) . Membranes were blocked in 5% milk with 0 . 1% Tween 20 and subsequently incubated with monoclonal antibody 13C9 ( DSHB , 1:20 ) , rabbit anti-HA ( Covance , 1:1000 ) or anti-C-Myc ( Santa Cruz , 9E10 1:250 ) ( to confirm increasing levels of PRRG4 protein ) . Proteins were detected using HRP-conjugated secondary antibodies ( Jackson Laboratories , 1:5000 ) and visualized with ECL detection reagents in a ChemiDoc imager ( Bio-Rad ) . Signal intensities were measured in ImageJ . | Mutants for the fruit fly commmissureless gene ( comm ) dramatically lack connections between the left and right hand sides of the nervous system . This is due to a failure to prevent Robo receptors from reaching the cell surface , where they guide growing axons away from the CNS midline . Comm proteins are not thought to exist outside of insects . By carefully comparing proteins from other species , candidate homologues from vertebrates and yeast were identified . The candidates were tested by expression in the fly nervous system and one gene , PRRG4 , was found to affect the phenotype caused by expression of the human Robo1 gene . When Robo genes are expressed in cell culture , they localize to the surface of the cell . PRRG4 was found to be able to re-localize Robo away from the cell surface , a property shared with Comm protein , indicating that they are functional homologues . Human patients with WAGR syndrome often display autistic features and these have been attributed to loss of one copy of PRRG4 . Our findings suggest that PRRG4 guides growing axons and that brain wiring patterns may be subtly altered in WAGR patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"pervasive",
"developmental",
"disorders",
"autism",
"social",
"sciences",
"neuroscience",
"animals",
"developmental",
"psychology",
"animal",
"models",
"membrane",
"proteins",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"nerve",
"fibers",
"sequence",
"motif",
"analysis",
"embryos",
"autism",
"spectrum",
"disorder",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"embryology",
"developmental",
"neuroscience",
"neurodevelopmental",
"disorders",
"bioinformatics",
"animal",
"cells",
"axons",
"insects",
"cell",
"membranes",
"arthropoda",
"cellular",
"neuroscience",
"psychology",
"cell",
"biology",
"anatomy",
"central",
"nervous",
"system",
"axon",
"guidance",
"database",
"and",
"informatics",
"methods",
"neurons",
"neurology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2017 | The WAGR syndrome gene PRRG4 is a functional homologue of the commissureless axon guidance gene |
Phenotypic plasticity is the ability of a single genotype to produce different phenotypes in response to changing environments . We assessed variation in genome-wide gene expression and four fitness-related phenotypes of an outbred Drosophila melanogaster population under 20 different physiological , social , nutritional , chemical , and physical environments; and we compared the phenotypically plastic transcripts to genetically variable transcripts in a single environment . The environmentally sensitive transcriptome consists of two transcript categories , which comprise ∼15% of expressed transcripts . Class I transcripts are genetically variable and associated with detoxification , metabolism , proteolysis , heat shock proteins , and transcriptional regulation . Class II transcripts have low genetic variance and show sexually dimorphic expression enriched for reproductive functions . Clustering analysis of Class I transcripts reveals a fragmented modular organization and distinct environmentally responsive transcriptional signatures for the four fitness-related traits . Our analysis suggests that a restricted environmentally responsive segment of the transcriptome preserves the balance between phenotypic plasticity and environmental canalization .
Phenotypic plasticity is the ability of a single genotype to give rise to different phenotypes in different environments [1] . Phenotypic plasticity is the counterpoint to environmental canalization [2]–[3] , whereby genotypes produce the same phenotype in different environments . Phenotypic plasticity allows organisms to respond rapidly to changing environmental conditions without the time lag required for response to natural selection on segregating allelic variants and without the cost of selection , while environmental canalization buffers phenotypes against environmental perturbations . The balance between plasticity and robustness is thus crucial for optimal fitness [3]–[4] in variable environments , but the genetic basis for phenotypic plasticity has remained poorly defined . Elucidating the genetic underpinnings of phenotypic plasticity ( and its converse , environmental canalization ) requires that we determine what fraction of the genome is environmentally sensitive , which genes respond to the same or different environmental perturbations and how expression of environmentally sensitive genes is correlated with plasticity of organismal phenotypes . It is also necessary to determine what the relationship is between genetic variance and phenotypic plasticity , whether the same genes affecting phenotypic plasticity for a trait also affect genetic variation for that trait , and whether environmentally plastic and environmentally robust genes evolve at different rates . Although previous studies have analyzed changes in gene expression under one or few different environmental or physiological conditions [5]–[11] , the study presented here is the first comprehensive study that analyzes co-variation among environmentally responsive genes across a wide range of environments in a defined outbred population reconstructed from inbred lines with documented expression profiles , enabling us to compare genotypic and environmental variation . We examined phenotypic plasticity in genome-wide gene expression and four organismal phenotypes related to reproductive fitness in a population generated by crossing 40 wild-derived inbred D . melanogaster lines [12] . The majority of the transcriptome shows robust expression across a range of environmental challenges , including different nutritional rearing conditions , physical stress conditions , chemical exposures , social crowding during larval or adult stages , and aging . Approximately 15% of transcripts are phenotypically plastic . By comparing genotypic variation among the original 40 wild-derived inbred lines under standard growth conditions , documented earlier [12] , with environmental variation of transcript abundance levels in the reconstituted outbred population , we were able to discriminate two distinct classes of environmentally responsive transcripts , which we have designated Class I and Class II transcripts .
To identify phenotypically plastic and environmentally canalized transcripts , we assessed genome-wide gene expression of flies exposed to 20 treatments , including a control treatment of mated flies reared under standard conditions , and different nutrient or drug supplements , exposure to different physical and social environments , and maintenance at different reproductive states . Of the 18 , 800 transcripts represented on the microarray , 14 , 400 ( 76 . 6% ) generated signal intensities above background under at least one treatment , similar to the proportion of the transcriptome detected in a previous study , in which transcript profiles were obtained separately for the 40 individual genotypes that gave rise to our outbred population [12] . Analysis of variance of microarray intensity signals across all 20 rearing conditions revealed 1 , 249 transcripts that showed a significant treatment effect ( 8 . 7% ) , 6 , 745 transcripts that showed a sex effect ( 46 . 8% ) , and 200 transcripts with a significant treatment by sex interaction term ( 1 . 4% ) at a false discovery rate of 0 . 05 . Thus , the majority of the transcriptome is remarkably robust and buffered against diverse environmental challenges . We refer to the 1 , 249 transcripts exhibiting phenotypic plasticity as quantified by the significant treatment term in the ANOVAs as Class I transcripts . To simplify statistical analyses and maintain optimum power we excluded 166 Class I transcripts that also had significant treatment by sex interaction terms , giving 1 , 133 phenotypically plastic transcripts for further analyses ( Table S1 ) . We grouped these transcripts according to their relative expression levels across the 20 conditions ( Figure S1 ) . The highest relative gene expression levels were observed in flies exposed to low temperature , dopamine , nicotine or high sugar , and the lowest relative levels in flies exposed to heat shock or grown on high yeast medium . Surprisingly , overall relative gene expression levels are either uniformly higher or lower ( more than 70% show higher abundance levels than the median ) across the 20 conditions; however , starvation stress resistance , aging , larval crowding and exposure to high temperature , result in substantial variation among relative expression levels ( Figure S1 ) . To further examine the relationship between gene expression and environmental exposure , we compared transcript abundance levels of the Class I transcripts under the different treatments to the standard rearing condition with post hoc least square difference ( LSD ) tests ( p<0 . 05; Table S2 ) . Heat shock has the greatest impact on gene expression ( 589 transcripts ) , whereas fluoxetine changes expression of only four transcripts ( Figure 1A ) . Many transcripts show altered expression under multiple treatments; for example , 167 transcripts show altered expression both after heat shock and exposure to starvation ( Figure 1A ) . The majority of transcripts do not undergo significant changes compared to the standard condition ( Figure 1B ) . Among the 1 , 133 Class I transcripts , 691 are computationally predicted with unknown function , 14 probe sets correspond to intergenic regions , non-coding RNAs and transposons , and 428 are annotated . The transcripts that show altered expression after heat shock include 13 Heat shock proteins ( Hsps ) , 60 proteases , 17 members of the cytochrome P450 family ( Cyps ) , two glutathione-S-transferases ( Gsts ) , six UDP-glucose glycoprotein glycosyltransferases ( Ugts ) , and six immune-induced molecules ( IM ) ( Figure 2 , Table S2; Figure S2 ) . A variety of additional transcripts in diverse gene ontology ( GO ) categories also respond to environmental challenges ( Table S3 ) . Nine of the 13 Hsps upregulated after heat shock are also upregulated after induction of chill coma ( Figure S2 ) . The abundance of heat shock proteins and proteases encoded by environmentally sensitive genes reflect mechanisms for environmental adaptation of the proteome . Heat shock proteins may offer protection for nascent polypeptides under adverse temperature conditions , while one function of environmentally sensitive proteases may be to facilitate de novo protein synthesis by providing a pool of amino acids through degradation of dispensable proteins . We asked to what extent expression patterns of phenotypically plastic transcripts are co-regulated across environmental treatments . A previous study on the 40 inbred lines from which our outbred population is derived demonstrated that the genetically variable transcriptome ( 10 , 096 transcripts ) is highly inter-correlated and can be subdivided into 241 co-regulated modules [12] , identified by modulated modularity clustering ( MMC ) [12]–[13] . Here , we used MMC to identify transcripts that covary across different treatments ( Figure 3A , Table S1 ) . This analysis partitioned the 1 , 133 Class I transcripts into 103 small , but highly correlated transcriptional modules ( the average absolute correlation coefficient , |r| , within modules is at least 0 . 56 ) . Extensive cross-module correlations are also evident . Negative correlations are rare , in agreement with the overall uniform up- or down-regulation of transcripts ( Figure S1 ) . All seven IM transcripts group together in module 71 . A putative IM , CG15065 , which is genetically correlated with IM1 and IM3 [12] , is also contained in this module ( Figure S2 ) . These results show that changes in environmental conditions can cause fragmentation of the highly intercorrelated structure of the transcriptome observed under the standard growth condition [12] . What are the cellular mechanisms that regulate transcriptional responses to environmental changes ? As a first step to investigating how environmental stimuli may influence transcriptional regulation , we asked which transcription factors show altered expression under the different environmental conditions . Among the Class I transcripts , we identified 26 transcripts that encode transcriptional regulators , of which 25 were differentially expressed relative to the standard growth condition ( Figure 3B , Figure 4A ) . Many of these transcription factors occur together in the same transcriptional modules ( Figure 3B ) . The complexity of the interrelationships between transcript abundance levels of different transcription factors under different growth conditions is further illustrated in Figure 4A . Each transcription factor can exert wide-ranging effects on networks of interacting genes that include other regulatory genes , non-regulatory genes and miRNAs [14]–[15] ( Figure 4B ) . Direct genetic , protein-protein and gene-protein interactions among these phenotypically plastic transcription factors appear compartmentalized , with little overlap between interacting components related to each transcription factor . In contrast , there is an extensive network of interactions among microRNAs and different transcription factors ( Figure 4B ) . Extensive interactions of transcription factors with miRNAs suggest that these may also contribute to phenotypic plasticity of the transcriptome [14]–[15] . In addition , seven long non-coding RNAs and unannotated intergenic regions are phenotypically plastic , and genes that encode several phenotypically plastic transcripts contain overlapping or flanking sequences for short non-coding RNAs ( Table S1 ) . Finally , we note that predicted transcripts of unknown function may also play a regulatory role . In addition , transcriptional regulators that themselves show no change in gene expression may be regulated by phenotypically plastic post-translational modifications . We next asked to what extent the phenotypic plasticity in gene expression is associated with phenotypic plasticity of organismal phenotypes . We assessed phenotypic plasticity of four fitness-related phenotypes: development time , lifespan , starvation stress resistance , and chill coma recovery time . Development is exquisitely sensitive to environmental conditions [16]–[17] ( Figure 5A ) and is accelerated when flies are grown on medium supplemented with high yeast , and delayed when the medium is supplemented with high sugar . When grown on both high sugar and high yeast , development time is identical to that under the standard growth condition . Growth at 28°C also accelerates development , but reduces survival , whereas growth at 18°C delays development about 2-fold . Larval crowding and exposure to the chemical oxidative stress agent menadione sodium bisulfite results in delayed development along with reduced survival . Medium supplemented with 10% ethanol has a small effect on development time and survival . All other treatments lead to slower development compared to the standard condition . In addition to prolonging development , growth at 18°C results in a two-fold increase in lifespan ( Figure 5B ) . Furthermore , virgin females live longer than mated females , as expected [18] . When reared under standard conditions , subsequent food deprivation allows females to survive twice as long as males ( Figure 5C ) . Survival curves indicate a trend towards longer survival times for both sexes when flies are maintained at high density , either as larvae or adults , or at 18°C . Exposure to chill coma tends to increase survival time during subsequent food deprivation in females ( Figure 5C ) . Previous exposure to heat shock extends chill coma recovery time . Recovery is also delayed as a result of aging , growth at 28°C , previous exposure to starvation stress , growth on ethanol , menadione sodium bisulfite , or high sugar-supplemented medium , and maintenance at high density as adults ( Figure 5D ) . Whereas caloric restriction extends lifespan [19]–[20] , a single 24 h food deprivation period does not affect lifespan . The increase in starvation stress resistance following chill coma recovery may be due to slowing of intermediary metabolism during chill coma and , consequently , preservation of metabolic energy . We used regression to identify Class I transcripts associated with variation in organismal phenotypes across the 20 environmental conditions , and MMC to construct environmentally correlated modules [13] of these transcripts ( Figure 6 ) . Phenotypic plasticity in development time is associated with 426 transcripts , of which 411 cluster into 36 highly correlated ( average |r|>0 . 5 ) modules ( Figure 6A ) . Similarly , phenotypic plasticity in lifespan , starvation stress resistance and chill coma recovery is associated with , respectively , 186 , 320 and 440 transcripts , which cluster into 12 , 30 and 23 highly correlated ( average |r|>0 . 5 ) modules , respectively ( Figure 6B–6D , Tables S4 , S5 , S6 , S7 ) . Modules associated with each trait show high degrees of inter-correlation , and there is evidence for cross-module clustering , indicating hierarchical co-regulation of the Class I plastic genes ( Figure 6 ) . We found little overlap ( ∼3% ) between transcripts associated with genetic variation in lifespan , starvation resistance , and chill coma recovery under the standard growth condition [12] and transcripts associated with phenotypic plasticity of these traits . Since 1 , 125 of the 1 , 133 ( 99 . 3% ) Class I transcripts are also genetically variable , the lack of concordance between the association with genetic and environmental phenotypic variation cannot be attributed to the trivial explanation that the genetically variable and phenotypically plastic transcripts do not overlap . Some modules associated with different organismal phenotypes are enriched for common transcripts , indicating pleiotropy for phenotypic plasticity ( Figure S3 ) . For example , the chill coma recovery module 17 contains the same transcripts as modules associated with phenotypic plasticity in development time and in starvation stress resistance . Whereas pleiotropy at the level of covariant transcriptional modules is prominent between chill coma recovery , starvation stress resistance and development time , it is sparser between lifespan and the other traits ( Figure S3 ) . In summary , clustering analysis of Class I transcripts reveals a fragmented modular organization and distinct environmentally-responsive transcriptional signatures for the four fitness-related traits . To assess the relationship between genetic variation and phenotypic plasticity , we compared the previously reported genetic variance and micro-environmental variation ( within-line variation ) across the 40 inbred lines [12] from which our outbred population was derived , reared under the standard growth condition , with the variation in phenotypic plasticity ( macro-environmental variation ) and within-treatment variation of the same transcripts in the outbred population . We quantified genetic variation as the coefficient of variation among lines ( CVL ) and variation in phenotypic plasticity as the coefficient of macroenvironmental variance ( CVME ) . We found a strong correlation between genetic variation and variation in phenotypic plasticity for Class I transcripts in both sexes ( Figure 7A , 7B ) . This comparison revealed an additional group of 982 environmentally sensitive transcripts with high macroenvironmental variation , but low genetic variance ( Figure 7A , 7B , Figure S4 , Table S1 ) . These phenotypically plastic transcripts , which we designate as Class II , were not identified by our initial analysis due to high within-treatment environmental variation ( quantified as the coefficient of variation within environments , CVEW , Figure S5A–S5D ) . Phenotypic plasticity for Class II transcripts was mostly sexually dimorphic , with 230 transcripts specific to females , 560 specific to males , and 192 in common for both sexes ( Table S1 ) . Class I transcripts have greater genetic variation for both sexes than environmentally robust transcripts , which are relatively stably expressed both across genotypes and treatments ( Figure 7G , 7H , Figure S4C , S4D ) . In males the average genetic variance of Class II transcripts is lower than both Class I and robust transcripts ( Figure 7G , Figure S4A ) , while in females the average genetic variance of Class II transcripts is lower than the Class I but higher than the robust transcripts ( Figure 7H , Figure S4B ) . In contrast to the genetic variance , the average macroenvironmental variation across treatments of Class II transcripts is ∼two-fold greater than that of Class I transcripts for both sexes ( Figure 7G , 7H , Figure S4C , S4D ) . There is little correlation between the level of genetic ( Figure 7C , 7D ) and macroenvironmental ( Figure 7E , 7F ) variation with the mean level of gene expression across all environments for Class I and robust transcripts . However , the macroenvironmental variance ( Figure 7E , 7F ) as well as the variance in gene expression within treatments ( Figure S5A–S5D ) and within inbred lines ( Figure S5E , S5F ) are correlated with the mean expression across treatments . To exclude the possibility that this observation is an artifact due to array quality , we examined the correlation between the previously published mean expression levels of transcripts across the 40 inbred lines [12] and the mean expression level in the outbred population across conditions . Transcript means were highly correlated between the two experiments ( r = 0 . 960 and r = 0 . 963 , for females and males , respectively; Figure S5G , S5H ) . Since Class II transcripts exhibited sexual dimorphism in phenotypic plasticity , we evaluated the relationship between sexual dimorphism in mean gene expression across all 20 environments , and sexual dimorphism for phenotypic plasticity , for Class I and Class II transcripts as well as a sample of robust transcripts ( Figure S6 ) . We found a clear inverse relationship between sexual dimorphism for mean expression and sexual dimorphism for the variance in expression across environments for Class I transcripts , but not the other categories . Female-biased Class II genes for mean expression are male-biased for plasticity in expression , and vice versa . Class II phenotypically plastic transcripts can be further classified into high and low expression categories . Highly expressed transcripts in females overlap transcripts with low expression in males , and GO analysis shows that these 19 transcripts encode yolk proteins and chorion proteins and are enriched for oogenesis and sexual reproduction ( Table S8 ) . Similarly , transcripts with low expression in females overlap transcripts with high expression in males , and GO analysis shows that these 162 transcripts encode male-specific proteins , accessory gland proteins and hormones which affect mating and post-mating behaviors ( Table S9 ) . Further GO analyses indicate that female-specific Class II transcripts are enriched in mitochondria- and muscle-related functions , whereas male-specific transcripts are enriched in functions of cuticular structure and DNA replication in meiosis ( Tables S10 and S11 ) . Enrichment of Class II transcripts for reproductive functions suggests that the high environmental responsiveness of these transcripts may protect reproductive fitness . To assess to what extent phenotypically plastic genes are evolutionarily conserved compared to the rest of the genome , we looked at the percentage of homologues across 12 Drosophila species , the ratio of non-synonymous to synonymous substitutions ( ω ) and fraction of adaptive fixations ( α ) using D . yakuba as outgroup [21]–[23] ( Figure 8 , Figure S7 ) . Class I genes are less conserved across the Drosophila clade and have less constraints under selection than the environmentally robust genome . Previously , we documented plasticity of the Drosophila chemoreceptor repertoire [24] . Like chemoreceptor genes , many of the Class I transcripts also belong to rapidly evolving multigene families ( Figure 8 ) . Such rapid evolution may involve gene duplication and subfunctionalization , as is evident within the Cyp gene family [25]–[29] . Class II genes show an even faster rate of evolution compared to robust transcripts with a significantly higher proportion of positively selected sites , as evident from the distributions of ω and α ( Figure 8 ) . Thus , these phenotypically plastic genes appear especially responsive to natural selection .
Genome-wide transcriptional analysis of flies reared under 20 environmental conditions shows that ∼15% of the transcriptome exhibits phenotypic plasticity , while the rest is environmentally canalized . Logistical and economic constraints have limited this initial investigation to whole flies . We surveyed the FlyAtlas database [30] and found that transcripts associated with all organismal phenotypes are generally expressed in multiple , but not all tissues ( Figure S8 ) . In future studies it would be of interest to examine directly tissue-specific environmental modulation of expression of phenotypically plastic transcripts . Since we only examined adult flies , we could not detect transcripts that may show environmental plasticity at different developmental stages . Furthermore , although we provided a comprehensive analysis of the transcriptional response to a vast variety of conditions and treatments , additional treatments , e . g . different chemical exposures or sleep deprivation , might reveal additional features of the phenotypically plastic transcriptome . However , results from previous studies on genome-wide transcriptional responses to environmental and physiological changes in Drosophila are in line with our observations [6] , [8] , [11] . A previous study that examined phenotypic plasticity of the transcriptome during aging and upon exposure to paraquat-induced oxidative stress reported altered regulation of Hsp26 , several metabolic enzymes , and glutathione-S-transferases [6] . Genome-wide transcriptional profiling during aging under conditions of caloric restriction also showed plastic responses of genes associated with xenobiotic defense and reproduction [8] . In addition , phenotypically plastic transcripts associated with xenobiotic defense , metabolism and chitin biosynthesis have been identified in Drosophila populations from tropical and temperate zones in Eastern Australia in response to temperature [11] . The analysis presented here is the most comprehensive study of phenotypic plasticity to date , which capitalizes on the unique properties of an outbred population reconstructed from well characterized inbred wild-derived lines , which enabled us to discriminate two classes of plastic transcripts . Class I transcripts are not only phenotypically plastic , but are more genetically variable and evolve more rapidly than the rest of the transcriptome . Class I transcripts are enriched in functions of detoxification , metabolism , proteolysis and heat shock proteins . Class I transcripts also encode gene products of unknown function , including non-coding RNAs , which may contribute to modulation of chromatin structure and transcriptional regulation . The coupling of high genetic variation within a population and rapid evolution suggests interesting evolutionary forces acting on these genes . Class II transcripts have low genetic variance for mean expression levels , but greater environmental variation in transcript abundance , and are even more rapidly evolving than Class I transcripts . It is tempting to speculate that reduced genetic variation for these transcripts within a population is the consequence of selection favoring genotypes with high phenotypic plasticity within each species , but with variable selection pressures across species [31] . Under this hypothesis genotypes with high transcriptional plasticity would be fixed within a species but divergent across species , which implies there is genetic variation in phenotypic plasticity on which selection acts . We note , however , that non-additive effects may confound inferences based on comparing an outbred population in many environments with inbred genotypes in one environment . Two models of the genetic basis of phenotypic plasticity have been postulated [32] . Under the ‘allelic sensitivity’ model , the same alleles affect the mean value of a phenotype and its plastic response to environmental variation . Under the ‘gene regulation’ model , plasticity is a trait in itself , under the control of regulatory loci which modulate the expression of other genes in different environments . Our comparison of transcripts for phenotypic plasticity in an outbred population with genetically variable transcripts among the 40 inbred genotypes from which the outbred population was derived support the gene regulation hypothesis . We found no more overlap than expected by chance between transcripts associated with the mean and plasticity of four fitness-related traits . Class II phenotypically plastic transcripts are highly sexually dimorphic , but male-biased plastic transcripts are associated with female-biased mean expression levels , and vice versa , again suggesting an uncoupling between the mean and macroenvironmental variance . The Class I plastic transcripts cluster into modules of highly correlated transcripts , with a high degree of correlation across modules , further implicating co-regulation of plastic responses to environmental variation . Whereas this initial comprehensive survey of phenotypic plasticity is necessarily largely descriptive , it provides a foundation for future studies aimed at testing mechanisms that link environmental inputs to alterations in gene expression . It is likely that the environmentally plastic transcriptional regulators which we identified ( Figure 4 ) will play a role in mediating these responses . Furthermore , since we did not consider the effects of DNA sequence variants on phenotypic plasticity , co-regulated modules of phenotypically plastic transcripts are undirected . Deriving causal transcriptional networks for genetic variation in phenotypic plasticity requires superimposing genetic variation [33] . The recent availability of whole genome sequences of the wild-derived inbred lines from the Raleigh population will enable such analyses in the future [23] .
We generated a synthetic outbred population by round-robin crossing of 40 wild derived inbred lines of the Drosophila Genetic Reference Panel ( DGRP ) [12] , [23] , followed by random mating for over 47 generations . For age-synchronization , we randomly collected 1000 females and 1000 males and allowed oviposition for 12 h on grape agar plates supplemented with yeast paste . Unless indicated otherwise , 55 eggs were collected and subjected to different treatments throughout development . The standard rearing condition ( cornmeal ( 65 g/L ) -molasses ( 45 ml/L ) -yeast ( 13 g/L ) - agar medium at 25°C , 60–75% relative humidity and a 12 h light-dark cycle ) resulted in hatching of ∼50 larvae . Adults were collected immediately after eclosion , and placed at a density of 25 females and 25 males under the desired condition . Flies were transferred onto fresh medium every two days . For nutritional and pharmacological treatments , flies were reared on standard medium supplemented with 225 ml/L molasses ( ‘high sugar’ ) , 65 g/L yeast ( ‘high yeast’ ) , 225 ml/L molasses and 65 g/L yeast ( ‘high sugar-high yeast’ ) , 10% ( v/v ) ethanol , 200 µM fluoxetine hydrochloride , 47 mM dopamine , 1 mM nicotine , 2 mM caffeine or 4 mM menadione sodium bisulfite . Different physical environments included constant light , 28°C ( ‘high temperature’ ) , 18°C ( ‘low temperature’ ) , and exposure to different stresses , including heat shock ( 37°C for 1 h; 1 h recovery prior to RNA extraction ) , chill coma ( 3 h on ice; 1 h recovery prior to RNA extraction ) , and 24 h starvation . Different social environments included larval crowding ( 300 eggs/vial ) and adult crowding ( 80 females and 80 males were pooled in each vial immediately after eclosion ) . To compare mated with non-mated flies , 50 single sex virgins were reared separately . Flies reared under standard conditions were mated . Aged flies were 30 days old . We used Affymetrix Drosophila 2 . 0 arrays to assess whole genome transcriptional profiles . Males and females ( 3–5 days old ) were collected between 1:00–3:00 pm by aspiration and immediately frozen on dry ice . RNA was extracted from three independent samples ( 30 flies/sex/condition ) , and 10 µg of biotinylated , fragmented cRNA was hybridized to each microarray . RNA extraction , labeling and hybridization were randomized across samples . Raw data were log2 transformed and normalized across sexes and conditions using a median standardization . For each probe set , we used the median log2 signal intensity as the measurement of expression . We used negative control probe sets to estimate background intensity . Probe sets with hybridization intensities below background under all different treatment conditions were removed from the analysis . We did not correct for probe mismatches due to segregating polymorphisms in the reconstituted outbred population , because ( 1 ) the average hybridization bias will be identical across all environmental conditions , and; ( 2 ) only about 3 , 000 single feature polymorphisms ( SFPs ) were identified among the original 40 inbred lines previously and their removal from the data set did not significantly influence the hybridization results [12] . Microarray data have been deposited in the ArrayExpress database ( accession: E-MTAB-639 and are also available on the DGRP website ( http://dgrp . gnets . ncsu . edu/ ) . We analyzed array data using a Generalized Linear Model ( GLM ) in SAS to partition phenotypic variation between sexes ( S , fixed ) , environments/treatments ( E , fixed ) , the S×E interaction ( fixed ) and the error variance ( ε ) . To identify environmentally responsive Class I transcripts we used an FDR<0 . 05 to correct for multiple tests . Post-hoc LSD tests were used to identify transcripts with a significant environment term . Signal intensities for those transcripts were sex-centered by subtracting the female or male mean across all conditions for each gene . Transcripts with a significant interaction term were excluded . To resolve Class II transcripts , we applied two filters . First , we selected transcripts with cross-treatment ( macroenvironmental ) variance >95th percentile of the macoenvironmental variance distribution of the Class I transcripts ( coefficients of variation across treatments >7 . 06 and >7 . 12 for females and males , respectively ) . We filtered these transcripts further using an FDR>0 . 0001 for genetic variation among DGRP lines [12] for females and males separately . Finally , we removed overlapping transcripts between Class I and Class II . We used a form of K-means clustering ( K = 2 ) to partition the Class II transcripts into groups of high and low expression . Specifically , for each sex we exhaustively identified the unique bipartition of Class II transcripts that minimized the total within group sum-of-squares . The Modulated Modularity Clustering ( MMC ) algorithm [13] was used to group transcripts into covariant modules . Gene annotations were based on Flybase , version 5 . 36 . Gene ontology analysis was done using the DAVID bioinformatics database , using the Benjamini correction of p<0 . 05 as criterion for enrichment [34] , [35] . To measure development time , we allowed flies to lay eggs for 3 hours ( 10:00am–1:00pm ) , after which 55 eggs were collected and placed under 14 different growth conditions ( 300 eggs were collected to assess development time under the larval crowding condition ) . We counted flies , sexes separately , that eclosed every 12 hours ( N = 4 vials/condition ) . Life span was measured by collecting three females and three males immediately 1–3 days after eclosion , transferring them to fresh vials every 2–3 days , and recording survival daily ( N = 26 replicates/condition ) . To measure starvation stress resistance , we placed ten 3–5 days old flies in vials containing 1 . 5% agar , and scored survival every 8 hours ( N = 4×10/sex/condition ) . To measure chill coma recovery , we placed 3–7 day-old flies in empty vials on ice for 3 hours , and determined their subsequent recovery time at room temperature by their ability to recover upright posture ( N = 2×50 flies/sex/condition ) . Phenotypic data are available on the DGRP website ( http://dgrp . gnets . ncsu . edu/ ) . We used regression to identify transcripts with variation in expression levels that associated with organismal phenotypic variation ( p<0 . 05 ) . For traits with a significant sex by environment interaction , regression was applied to sexes separately ( Y = μ+Exp+ε , where Exp denotes the covariate median log2 expression level ) . For traits without a significant sex by environment interaction , we used sex-centered measures of deviations from female or male means for both expression and organismal phenotypes . We used the residuals from the regressions ( Y = μ+T+ε , where T denotes the trait covariate ) to compute environmental correlations between transcripts significantly associated with each organismal phenotype for construction of covariant modules using MMC [13] . | Unlike Mendelian traits , where the genotype allows a direct prediction of the phenotype , predicting phenotypic values is not straightforward for complex traits , which arise from multiple segregating genes and their interactions with the environment . Here , a single genotype can often express different phenotypes in different environments . Such phenotypic plasticity is the counterpoint to “environmental canalization , ” whereby genotypes produce the same phenotype in different environments . Whereas phenotypic plasticity allows organisms to respond rapidly to changing environments , environmental canalization buffers phenotypes against environmental perturbations . The balance between plasticity and robustness is crucial for optimal fitness , but the genetic basis for phenotypic plasticity is poorly defined . Here , we present the most comprehensive analysis to date of variation in genome-wide gene expression of an outbred Drosophila melanogaster population under 20 different environments . We find that a restricted environmentally responsive segment of the transcriptome ( ∼15% ) preserves the balance between phenotypic plasticity and environmental canalization . Environmentally plastic transcripts can be grouped into two categories . Class I transcripts are genetically variable and associated with detoxification , metabolism , proteolysis , heat shock proteins , and transcriptional regulation . Class II transcripts have low genetic variance and show sexually dimorphic expression enriched for reproductive functions . Despite low genetic variance these transcripts evolve rapidly . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"model",
"organisms",
"genetics",
"biology",
"genomics",
"genetics",
"and",
"genomics"
] | 2012 | Phenotypic Plasticity of the Drosophila Transcriptome |
Sensory information about the outside world is encoded by neurons in sequences of discrete , identical pulses termed action potentials or spikes . There is persistent controversy about the extent to which the precise timing of these spikes is relevant to the function of the brain . We revisit this issue , using the motion-sensitive neurons of the fly visual system as a test case . Our experimental methods allow us to deliver more nearly natural visual stimuli , comparable to those which flies encounter in free , acrobatic flight . New mathematical methods allow us to draw more reliable conclusions about the information content of neural responses even when the set of possible responses is very large . We find that significant amounts of visual information are represented by details of the spike train at millisecond and sub-millisecond precision , even though the sensory input has a correlation time of ∼55 ms; different patterns of spike timing represent distinct motion trajectories , and the absolute timing of spikes points to particular features of these trajectories with high precision . Finally , the efficiency of our entropy estimator makes it possible to uncover features of neural coding relevant for natural visual stimuli: first , the system's information transmission rate varies with natural fluctuations in light intensity , resulting from varying cloud cover , such that marginal increases in information rate thus occur even when the individual photoreceptors are counting on the order of one million photons per second . Secondly , we see that the system exploits the relatively slow dynamics of the stimulus to remove coding redundancy and so generate a more efficient neural code .
Throughout the brain , information is represented by discrete electrical pulses termed action potentials or ‘spikes’ [1] . For decades there has been controversy about the extent to which the precise timing of these spikes is significant: Should we think of each spike arrival time as having meaning down to millisecond precision [2]–[5] , or does the brain only keep track of the number of spikes occurring in much larger windows of time ? Is precise timing relevant only in response to rapidly varying sensory stimuli , as in the auditory system [6] , or can the brain construct specific patterns of spikes with a time resolution much smaller than the time scales of the sensory and motor signals that these patterns represent [3] , [7] ? Here we address these issues using the motion-sensitive neurons of the fly visual system as a model [8] . We bring together new experimental methods for delivering truly naturalistic visual inputs [9] and new mathematical methods that allow us to draw more reliable inferences about the information content of spike trains [10]–[12] . We find that as we improve our time resolution for the analysis of spike trains from 2 ms down to a fraction of a millisecond we reveal nearly 30% more information about the trajectory of visual motion . The natural stimuli used in our experiments have essentially no power above 30 Hz , so that the precision of spike timing is not a necessary correlate of the stimulus bandwidth; instead the different patterns of precise spike timing represent subtly different trajectories chosen out of the stimulus ensemble . Further , despite the long correlation times of the sensory stimulus , segments of the neural response separated by ∼30 ms provide essentially independent information , suggesting that the neural code in this system achieves decorrelation [13] , [14] in the time domain , thereby enhancing the efficiency of the code on time scales relevant to behavior [15] .
Flies exhibit a wide variety of visually guided behaviors , of which perhaps the best known is the optomotor response , in which visual motion drives a compensating torque , stabilizing straight flight [16] . This system offers many advantages for the exploration of neural coding and computation: There is a small group of identified , wide-field motion-sensitive neurons [8] that provide an obligatory link in the process [17] , and it is possible to make very long , stable recordings from these neurons as well as to characterize in detail the signal and noise properties of the photoreceptors that provide the input data for the computation . In free flight , the trajectory of visual motion is determined largely by the fly's own motion through the world , and there is a large body of data on flight behavior under natural conditions [15] , [18]–[20] , offering us the opportunity to generate stimuli that approximate those experienced in nature . But the natural visual world of flies involves not only the enormous angular velocities associated with acrobatic flight; natural light intensities and the dynamic range of their variations are very large as well , and both of the fly's compound eyes are stimulated over more than 2π steradians . All of these features are difficult to replicate in the laboratory [21] . As an alternative , we have moved our experiments outside [9] , so that flies experience the scenes from the region in which they were caught . We recorded from a single motion-sensitive cell , H1 , while rotating the fly along trajectories modeled on published natural flight trajectories ( see Methods for details ) . We should note that for technical reasons , these stimuli do not contain natural translation , pitch , and roll components , which may have an effect on the H1 responses; for other approaches to the delivery of naturalistic stimuli in this system see [22] . A schematic of our experiment , and an example of the data we obtained , are shown in Figure 1 . We see qualitatively that the responses to natural stimuli are very reproducible , and we can point to specific features of the stimulus—such as reversals of motion direction—that generate individual spikes and interspike intervals with better than millisecond precision . The challenge is to quantify these observations: Do precise and reproducible patterns of spikes occur just at some isolated moments , or does looking at the spike train with higher time resolution generally provide more information about the visual input ? Precise spike timing endows each neuron with a huge “vocabulary” of responses [1] , [2] , but this potential advantage in coding capacity creates challenges for experimental investigation . If we look with a time resolution of τ = 1 ms , then in each bin of size τ we can see either zero or one spike; across the behaviorally relevant time scale of 30 ms [15] the neural response thus can be described as a 30-bit binary word , and there are 230 , or roughly one billion such words . Although some of these responses never occur ( because of refractoriness ) , and others are expected to occur only with low probability , it is clear that if precise timing is important then neurons can generate many more meaningfully distinguishable responses than the number that we can sample in realistic experiments . Can we make progress on assessing the information content and meaning of neural responses even when we can't sample all of them ? Recall that the information content is measured by the mutual information between the response and the stimulus that caused it [23] . This quantity measures ( in bits ) the reduction in the length of the description of the response spike train caused by knowing the associated velocity stimulus . Thus this mutual information is a difference of entropies [23] of the ensembles of all possible responses and the responses conditional on particular stimuli . Therefore , the problem of estimation of the information content of spike trains is essentially a problem of estimating the entropy of a probability distribution . This is known to be very hard when sampling is scarce , as in our problem [10] , [24] . Some hope is provided by the classical problem of how many people need to be present in a room before there is a reasonable chance ( about 50% ) that at least two of them share a birthday . This number , which turns out to be N∼23 , is vastly less than the number of possible birthdays , K = 365 . Turning this argument around , if we didn't know the number of possible birthdays we could estimate it by polling N people and checking the frequency of birthday coincidences . Once N is large enough to generate several coincidences we can get a pretty good estimate of K , and , for K→∞ , this happens when . Some years ago Ma proposed that this coincidence counting method be used to estimate the entropy of physical systems from molecular dynamics or Monte Carlo simulations [25] ( see also [26] ) . If these arguments could be generalized , it would become feasible to estimate the entropy and information content of neural responses even when experiments provide only a sparse sampling of these responses . The results of [10] , [11] provide such a generalization . To understand how the methods of [10] generate more accurate entropy estimates from small samples , it is useful to think about the simpler problem of flipping a coin under conditions where we don't know the probability p that it will come up heads . One strategy is to count the number of heads nH that we see after N flips , and identify p = nH/N; if we then use this “frequentist” or maximum likelihood estimate to compute the entropy of the underlying distribution , it is well known that we will underestimate the entropy systematically [24] , [27] , [28] . Alternatively , we could take a Bayesian approach and say that a priori all values of 0<p<1 are equally likely; the standard methods of Bayesian estimation then will generate a mean and an error bar for our estimate of the entropy given N observations . As shown in Figure 2 , this procedure actually leads to a systematic overestimate of the entropy in cases where the real entropy is not near its maximal value . More seriously , this systematic error is larger than the error bars that emerge from the Bayesian analysis , so we would be falsely confident in the wrong answer . Figure 2 also shows us that if we use a Bayesian approach with the a priori hypothesis that all values of the entropy , rather than p , are equally likely , then ( and as far as we know , only then ) we find estimates such that the systematic errors are comparable to or smaller than the error bars , even when we have seen only one sample . Thus the problem of systematic errors in entropy estimation is not , as one might have thought , the problem of not having seen all the possibilities; the problem rather is that seemingly natural and unbiased prior hypotheses about the nature of the underlying probabilities correspond to highly biased hypotheses about the entropy itself , and this problem gets much worse when we consider distributions over many alternatives . The strategy of [10] thus is to construct , at least approximately , a ‘flat prior’ on the entropy ( see Methods for details ) . The results of [12] demonstrate that this procedure actually works for both simulated and real spike trains , where ‘works’ means that we generate estimates that agree with the true entropy within error bars even when the number of samples is much smaller than the number of possible responses . As expected from the discussion of the birthday problem , what is required for reliable estimation is that the number of coincidences be significantly larger than one [11] . We note that this estimation method is substantially different from other recent approaches , such as [4] , [24] , [29] , [30] , and we discuss the differences in some detail in the Discussion . The tools described above allow us to estimate the entropy of neural responses . We first analyze a long experiment in which the fly experiences a continuous trajectory of motion with statistics modeled on those of natural flight trajectories ( Figure 3; see Methods for details ) . As shown in Figure 4A , we examine segments of the response of duration T , and we break these segments into discrete bins with time resolution τ . For sufficiently small τ , each bin either has one or zero spikes , and hence the response becomes a binary word with T/τ bits , while in the opposite limit we let τ = T , and then the response is the total number of spikes in a window of size T; for intermediate values of τ , the responses are multi-letter words , but with larger than binary alphabet when more than one spike can occur within a single bin . An interesting feature of these words is that they occur with a probability distribution similar to the distribution of words in English ( Zipf's law; Figure 4B ) . This Zipf-like behavior emerges only for T>20 ms , and was not observed in experiments with less natural , white noise stimuli [4] . With a fixed value of T , improving our time resolution ( smaller τ ) means that we distinguish more alternatives , increasing the “vocabulary” of the neuron . Mathematically this means that the entropy S ( T , τ ) of the neural responses is larger , corresponding to a potentially larger capacity for carrying information . This is shown quantitatively in Figure 4C , where we plot the entropy rate , S ( T , τ ) /T . The question of whether precise spike timing is important in the neural code is precisely the question of whether this capacity is used by the system to carry information [2] , [4] . To estimate the information content of the neural responses , we followed the strategy of [4] , [31] . The information content of the ‘words’ generated by the neuron is always less than the total size of the neural vocabulary because there is some randomness or noise in the association of words with sensory stimuli . To quantify this noise we choose a five second segment of the stimulus , and then repeat this stimulus 100 times . At each moment 0<t<5 s in the cycle of the repeated stimulus , we look across the one hundred trials to sample the different possible responses to the same input , and with the same mathematical methods as before , we use these samples to estimate the ‘noise entropy’ Sn ( T , τ|t ) in this ‘slice’ of responses . The information which the responses carry about the stimulus then is given by I ( T , τ ) = S ( T , τ ) −〈Sn ( T , τ|T ) 〉t , where 〈…〉t denotes an average over time t , which implicitly is an average over stimuli . It is convenient to express this as an information rate Rinfo ( T , τ ) = I ( T , τ ) /T , and this is what we show in Figure 4D , with T = 25 ms , chosen to reflect the time scale of behavioral decisions [15] . The striking feature of Figure 4D is the growth of information rate with time resolution . We emphasize that this measurement is made under conditions comparable to those which the fly encounters in nature—outdoors , in natural light , moving along trajectories with statistics similar to those observed in free flight . Thus under these conditions , we conclude that the fly's visual system carries information about motion in the timing of spikes down to sub-millisecond resolution . Quantitatively , information rates double as we increase our time resolution from τ = 25 ms to below a millisecond , and the final ∼30% of this increase occurs between τ = 2 ms and τ≤0 . 5 ms . In the behaviorally relevant time windows [15] , this 30% extra information corresponds to almost a full bit from this one cell , which would provide the fly with the ability to distinguish reliably among twice as many different motion trajectories . The information rate tells us how much we can learn about the sensory inputs by examining the neural response , but it doesn't tell us what we learn . In particular , we would like to make explicit the nature of the extra information that emerges as we increase our time resolution from τ = 2 ms to τ<1 ms . In other words , we should look at what additional features of the stimulus are encoded by finer spike timing . In the following we will present examples to highlight some of these features . We look at particular “words” in a segment of the neural response , as shown in Figure 5 , and then examine the motion trajectories that corresponded to these words [32] . For simplicity , we consider all responses that had two spikes in successive 2 ms bins , that is the binary pattern 11 when seen at τ = 2 ms resolution . When we improve our time resolution to τ = 0 . 2 ms , some of these responses turn out to be of the form 10000000000000000001 , while at the other extreme some of the responses have the two spikes essentially as close as possible given the refractory period , 00000100000000100000 . Remarkably , as we sweep through these subtly different patterns—which all have the same average spike arrival time but different interspike intervals—the average velocity trajectory changes form qualitatively , from a smooth “on” ( negative to positive velocity ) transition , to a prolonged period of positive velocity , to a more complex waveform with off and on transitions in succession . Examining more closely the distribution of waveforms conditional on the different responses , we conclude that these differences among mean waveforms are in fact discriminable . Thus , variations in interspike interval on the millisecond or sub-millisecond scale represent significantly different stimulus trajectories . A second axis along which we can study the nature of the extra information at high time resolution concerns the absolute timing of spikes . As an example , responses which at τ = 2 ms resolution are of the form 11 can be unpacked at τ = 0 . 2 ms resolution to give patterns ranging from 01000000001000000000 to 00000000010000000010 , all with the same interspike interval but with different absolute arrival times . As shown in Figure 5 , all of these responses code for motion trajectories with two zero crossings , but the times of these zero crossings shift as the spike arrival times shift . Thus , whereas the times between spikes represent the shape of the waveform , the absolute arrival time of the spikes marks , with some latency , the time at which a specific feature of the waveform occurs , in this case a zero crossing . Again we find that millisecond and sub-millisecond scale shifts generate discriminable differences . The idea that sub-millisecond timing of action potentials can carry significant information is not new , but the clearest evidence comes from systems in which the dynamics of the stimulus itself has significant sub-millisecond structure , as in hearing and electroreception [6] , [33] . For slow stimuli , the best recorded temporal precision is generally a few milliseconds , and is observed very early in the sensory processing [34] . Even for H1 , experiments demonstrating the importance of spike timing at the ∼2 ms level [4] , [35] could be criticized on the grounds that the stimuli had unnaturally rapid variations . It is thus important to emphasize that , in the experiments described here , H1 did not achieve millisecond precision simply because the input had a bandwidth of about a kiloHertz; in fact , the stimulus had a correlation time of ∼55 ms ( Figure 6 ) , and 99 . 9% of the stimulus power was contained below 30 Hz ( Figure 3F ) . We are not aware of previous results where sub-millisecond temporal precision has been explicitly shown to encode such slow stimuli . The long correlation time of these naturalistic stimuli also raises questions about redundancy—while each spike pattern considered in isolation may be highly informative , the long correlation time of the stimulus could very well mean that successive patterns carry information about essentially the same value of the instantaneous velocity . If so , that would mean that successive symbols are significantly redundant . Certainly on very short time scales this is true: Although Rinfo ( T , τ ) actually increases at small T since larger segments of the response reveal more informative patterns of several spikes [35] , [36] , it does decrease at larger T , a clear sign of redundancy . However , this approach to a constant information rate is very fast: We measure the redundancy on time scale T by computing YI ( T , τ ) = 2I ( T , τ ) / ( 2T , τ ) −1 , where YI = 0 signifies that successive windows of size T provide completely independent information , and YI = 1 that they are completely redundant . As shown in Figure 6 , YI ( T , τ ) decays rapidly , on a time scale of less than 20 ms . In contrast , correlations in the stimulus itself decay much more slowly , on the ∼55 ms time scale , and we find that the time dependent spike rate r ( t ) essentially has the same correlation time as the stimulus . The fact that coding redundancy decays three times more rapidly than the correlations of the time dependent firing rate indicates that the decorrelation of information is a process more intricate than simply filtering the stimulus . It suggests that there may be an adaptational mechanism at play that increases the overall efficiency of coding by exploiting the difference in time scales between stimulus changes and spike timing precision . If correct , this would imply that we should interpret neural firing patterns in context: The same pattern could signify slightly different stimulus depending on what went on before . This point merits further study , and may lead to further refinements in how we should interpret neural firing patterns , such as those shown in Figure 5 . As far as we know this is the first direct information theoretic demonstration of temporal redundancy reduction in the context of neural coding . The ability of the fly's visual system to mark features of the stimulus with millisecond precision , even at a ∼55 ms stimulus correlation time , was demonstrated in conditions where the visual input had very high signal-to-noise ratio . Previous work has suggested that this system can estimate motion with a precision close to the limits set by noise in the photoreceptors [37] , [38] , which is dominated by photon shot noise [39] , [40] . The present experiments , however , were done under very different conditions: Velocities of motion were much larger , the fly's eye was stimulated over a much larger area , and light intensities outdoors were much larger than generated by laboratory displays . Light intensities in our experiment were estimated to correspond to up to about 1 . 1·106 transduced photon/s per photoreceptor ( see Methods ) . Is it possible that photon counting statistics are limiting the precision of H1 , even at these high rates ? Because the experiments were done outdoors , there were small fluctuations in light intensity from trial to trial as clouds drifted by and obscured the sun . Although the range of these fluctuations was less than a factor two , the arrival times of individual spikes ( e . g . , the “first spike” after t = 1 . 75 s in Figure 1 ) had correlation coefficients of up to ρ = −0 . 42 with the light intensity , with the negative sign indicating that higher light intensities led to earlier spikes . One might see this effect as a failure of the system to adapt to the overall light intensity , but it also suggests that some of what we have called noise really represents a response to trial-by-trial variations in stimulus conditions . Indeed , a correlation between light intensity and spike time implies that the noise entropy Sn ( T , τ|t ) in windows which contain these spikes has a significant contribution from stimulus variation , and should thus be smaller when this source of variation is absent . More subtly , if photon shot noise is relevant , we expect that , on trials with higher light intensity , the neuron will actually convey more information about the trajectory of motion . We emphasize that this is a delicate question . To begin , the differences in light intensity were small , and we expect ( at most ) proportionately small effects . Further , as the light intensity increased , the total spike rate increased . Interestingly , this increased both the total entropy and the noise entropy . To see if the system used the more reliable signal at higher light intensities to convey more information , we have to determine which of these increases is larger . To test the effects of light intensity on information transmission ( see Methods for details ) , we divide the trials into halves based on the average light intensity over the trial , and we try to estimate the information rates in both halves; the two groups of trials differ by just 3% in their median light intensities . Since cutting the number of trials in half makes our sampling problems much worse , we focus on short segments of the response ( T = 6 ms ) at high time resolution ( τ = 0 . 2 ms ) ; note that these are still “words” with 30 letters . For this case we find that for the trials with higher light intensities the information about the motion stimulus is larger by Δ = 0 . 0204±0 . 0108 bits , which is small but significant at the 94% confidence level . We find differences with the same sign for all accessible combinations of T and τ , and the overall statistical significance of the difference thus is much larger . Note that since we were analyzing T = 6 ms windows , this difference correspond to ΔR∼3 bits/s , 1–2% of the total ( cf . Figure 4 ) . Thus even at rates of more than one million photons per second per receptor cell , small increases in photon flux produce proportionally small , yet measurable increases in the transmission of information about the motion stimulus .
We have found that under natural stimulus conditions the fly visual system generates spikes and interspike intervals with extraordinary temporal precision . As a consequence , the neural response carries a substantial amount of information that is available only at sub-millisecond time resolution . At this high resolution , absolute spike timing is informative about the time at which particular stimulus features occur , while different interspike intervals provide a rich representation of distinguishable stimulus features . These results clearly demonstrate that the visual system uses sub-millisecond timing to paint a more accurate picture of the natural sensory world , at least in this corner of the fly's brain . We emphasize again that here the sub-millisecond precision is not a result of an equally fast stimulus dynamics since the stimulus , in fact , has essentially no power at these frequencies . This is an important distinction , discussed in detail in [41] . In addition , an equally important observation is that the system performs efficiently both in the tasks of estimation and of coding , making use of the extra signal-to-noise provided by increased photon flux , even at daylight levels of light intensity . Perhaps of most interest , the analysis has made it possible to demonstrate a qualitative feature of the neural code in this system , namely the encoding of a temporally redundant stimulus in a neural signal of much shorter correlation time . At this point we can only speculate about the functional implications of this phenomenon , but at the very least it should give us pause in interpreting the code . Further study may reveal it to be an important feature of sensory coding and computation more generally , in particular under natural conditions where signals have high dynamic range , and show dramatic variations in reliability . We hope to be able to develop these ideas in more detail in the near future . Finally , we note that our ability to reach these conclusions depends not just on new experimental methods that allow us to generate truly naturalistic stimuli [9] , but critically on new mathematical methods that allow us to analyze neural responses quantitatively even when it was impossible for us to sample the distribution of responses exhaustively [10] , [12] . The theoretical tools presented here were developed with the explicit aim of being efficient in estimating entropies in the severely undersampled regime . This is crucial in neurophysiological experiments , where large stable datasets are very difficult to obtain . Most previously described entropy estimation methods , such as [4] , [24] , [27]–[30] , [42] , [43] , and others reviewed in [24] , have relied on one of three different ways to overcome the undersampling problem . Some , for example [29] , have chosen to define a metric on the space of responses , which makes it possible to “regularize” the problem by imposing similarity among probabilities of similar outcomes . Others , like [30] , explore generative models for the data , which serves a similar regularizing function . Both approaches work well if and only if the underlying choices match the properties of the real data . The majority of recent approaches , such as [24] , follow the third route and rely essentially on applying 1/N asymptotic corrections to the maximum likelihood estimator which means that they require mean bin occupancies O ( 1 ) to work . That leads to severe , and often impractical , demands on the size of the datasets as the cost of guaranteeing an estimator's performance . In contrast , the estimator presented here is based on counting coincidences , which still will occur even if the mean occupancy is much less than one . While we know that , in the worst case , even coincidence-based approaches may still require O ( 1 ) samples per possible outcome to produce low-bias and low-variance entropy estimates [44] , [24] , they may require substantially less data in simpler cases ( in the best case scenario , to reach equal levels of resolution , the number of independent samples in the data set scales as the square-root of the number required by the other estimation methods . Or alternatively , with the same size dataset , the timing resolution is better by a factor of two . ) For the data studied here , Nature cooperated: for example , to estimate noise entropies we use 100 samples for repeated stimuli for binary words of length 30 or more , so that the mean occupancy is <10−7 . However , the success of the method could not have been predicted a priori , and the majority of our computational effort was spent not on calculation of information rates per se , but on answering the very delicate question of whether the NSB method can be trusted to have small bias for our data . This is why we caution the reader from using NSB as a simple black-box estimation tool , without checking if it really works first . Finally , we notice that our method for estimating entropies bears some resemblance to the work of Wolpert and Wolf [45] , who used a single-beta Dirichlet prior to estimate functions of sparsely sampled probability distributions . A crucial distinction , however , is that instead of a single prior we use a family of Dirichlet priors to construct a prior distribution of entropies that is approximately flat ( see Methods ) . We believe that , without a similar flattening of the distribution of entropies , any Bayesian method is bound to have large biases below bin occupancies of O ( 1 ) . Information theoretic approaches force us to formulate questions and quantify observations in unbiased ways . Thus , success in solving a problem in an information theoretic context leads to results of great generality . But success in an experimental context hinges on the solution of practical problems . We hope that the methods presented here contribute to solving an important practical problem , and will be a step toward wider application of information theoretic methods in neuroscience .
H1 was recorded extracellularly by a short ( 12 mm shank length ) tungsten electrode ( FHC ) . The signal was preamplified by a differential bandpass instrumentation amplifier based on the INA111 integrated circuit ( Burr-Brown ) . After amplification by a second stage samples were digitized at 10 kHz by an AD converter ( National Instruments DAQCard-AI-16E-4 , mounted in a Fieldworks FW5066P ruggedized laptop ) . In off line analysis , the analog signal was digitally filtered by a template derived from the average spike waveform . Spikes were then time stamped by interpolating threshold crossing times . The ultimate precision of this procedure was limited by the signal to noise ratio in the recording; for typical conditions this error was estimated to be 50–100 µs . Note that we analyzed spike trains down to a precision of τ = 200 µs , so that some saturation of information at this high time resolution may have actually resulted from instrumental limitations . The experiments were performed outside in a wooded environment , with the fly mounted on a stepper motor with vertical axis . The speed of the stepper motor was under computer control , and could be set at 2 ms intervals . The DAQ card generated a 500 Hz clock signal divided down from the same master clock that governs the AD sample rate . The stepper motor ( SIG-Positec RDM566/50 , 10 , 000 pulses per revolution , or 0 . 036°/pulse ) was driven by a controller ( SIG-Positec Divistep D331 . 1 ) , which received pulses at a frequency divided down from a free running 8 MHz clock . Over the short time interval ( t , t+2 ms ) the stimulus velocity v ( t ) was determined by the pulse frequency , f ( t ) , that the controller received . This in turn was set by the numerical value , Ndiv ( t ) , of a divisor: f ( t ) = 8MHZ/Ndiv ( t ) , and v ( t ) = ( 0 . 036 ) · f ( t ) °/s . Successive values of Ndiv ( t ) were read every 2 ms from a stimulus file stored on a dedicated laptop computer . In this way , each 2 ms period the stepper motor speed was set to a value read from computer , keeping long-term synchrony with the data acquisition clock , with a maximum jitter of 1/ ( 8 MHz ) = 125 ns . The method for delivering pulses to the motor controller minimized the jerkiness of the motion by spacing the controller pulses evenly over each 2 ms interval . This proved to be crucial for maintaining stability of the electrophysiological recording . To stabilize temperature the setup was enclosed by a transparent plexiglass cylinder ( radius 15 cm , height 28 cm ) , with a transparent plexiglass lid . The air temperature in the experimental enclosure was regulated by a Peltier element fitted with heat vanes and fans on the inside and outside for efficient heat dispersal , and driven by a custom built feedback controller . The temperature was measured by a standard J-type thermocouple , and could be regulated over a range from some five degrees below to fifteen degrees above ambient temperature . The controller stabilized temperature over this range to within about a degree . In the experiments described here , temperature was 23±1°C . A running overall measure of light intensity was obtained by monitoring the current of a photodiode ( Hamamatsu S2386-44K ) enclosed in a diffusing ping pong ball . After a current to voltage conversion stage , the photodiode signal was amplified by a logarithmic amplifier ( Burr-Brown LOG100 ) operating over five decades . The probe was located ∼50 cm from the fly , and in the experiments the setup was always placed in the shade . The photodiode measurement was intended primarily to get a rough impression of relative light intensity fluctuations . To relate these measurements to outside light levels , at the start of each experiment a separate calibration measurement of zenith radiance was taken with a calibrated radiometer ( International Light IL1400A using silicon detector SEL033/F/R , with radiance barrel ) . The radiance measurement was done over a limited spectral band defined by a transmission filter ( International Light , WBS480 ) and an infrared absorption filter . In this way the radiometer's spectral sensitivity peaks close to the fly photoreceptor's 490 nm long wavelength maximum . However , it is about 20% broader than the fly's spectral sensitivity peak in the 350–600 nm range , and the photoreceptor's UV peak [46] was not included in this measurement . To relate this radiance measurement to fly physiology , the radiance reading was converted to an estimated effective fly photoreceptor photon rate , computed from the spectral sensitivity of the blowfly R1-6 type photoreceptor [46] , the radiometer's spectral sensitivity and the spectral distribution of sky radiance [47] . The reading of the photodiode was roughly proportional to the zenith intensity reading , with a proportionality factor determined by the placement of the setup and the time of day . In the experiments , light intensities within the visual field of the fly ranged from about 2% to 100% of zenith intensity . To obtain a practical rule of thumb , the photodiode readings were converted to equivalent zenith photon flux values , using the current to zenith radiance conversion factor established at the beginning of the experiment . During the experiments the photodiode signal was sampled at 1 s intervals . In their now classical experiments , Land and Collett measured the trajectories of flies in free flight [15]; in particular they reported the angular position ( orientation ) of the fly vs . time , from which we can compute the angular velocity v ( t ) . The short segments of individual trajectories shown in the published data have a net drift in angle , so we include both the measured v ( t ) and −v ( t ) as parts of the stimulus . We used the trajectories for the two different flies in Figure 4 of [15] , and grafted all four segments together , with some zero padding to avoid dramatic jumps in velocity , generating a 5 second long stimulus with zero drift , so that repetition of the angular velocity vs . time also repeated the angular position vs . time . Since Land and Collett reported data every 20 ms , we interpolated to generate a signal that drives the stepper motor at 2 ms resolution; interpolation was done using the MATLAB routine interp , which preserved the bandlimited nature of the original signal and hence did not distort the power spectrum . To analyze the full entropy of neural responses , it is useful to have a stimulus that is not repeated . We would like such a stimulus to match the statistical properties of natural stimulus segments described above . To do this , we estimated the probability distribution P[v ( t+Δt ) |v ( t ) ] from the published trajectories , where Δt = 20 ms was the time resolution , and then used this as the transition matrix of a Markov process from which we could generate arbitrarily long samples; our nonrepeated experiment was based on a 990 s trajectory drawn in this way . The resulting velocity trajectories , in particular , had exactly the same distributions of velocity and acceleration as in the observed free flight trajectories . Although the real trajectories are not exactly Markovian , our Markovian approximation also captures other features of the natural signals , for example generating a similar number of velocity reversals per second . Again we interpolated these trajectories to obtain a stimulus at 2 ms resolution . The problem in Figure 2 is that of a potentially biased coin . Heads appear with probability p , and the probability of observing n heads out of N flips is ( 1 ) If we observe n and try to infer p , we use Bayes' rule [1] to construct ( 2 ) where P ( p ) is our prior and is a normalization constant , which can be ignored . Given this posterior distribution of p we can calculate the distribution of the entropy , ( 3 ) We proceed as usual to define a function g ( S ) that is the inverse of S ( p ) , that is g ( S ( p ) ) = p; since p and 1-p give the same value of S , we choose 0<g≤0 . 5 and let g ˜ ( S ) = 1-g ( S ) . Then we have ( 4 ) From this distribution , we can estimate a mean S ˜N ( n ) and a variance σ2 ( n , N ) in the usual way . What interests us is the difference between S ˜N ( n ) and the true entropy S ( p ) associated with the actual value of p characterizing the coin; it makes sense to measure this difference in units of the standard deviation δS ( n , N ) . Thus we compute ( 5 ) and this is what is shown in Figure 2 . We consider two cases . First , a flat prior on p itself , so that P ( p ) = 1 . Second , a flat prior on the entropy , which corresponds to ( 6 ) Here , 1/2 in front of the derivative accounts for two values of p being mapped into the same S . Note that this prior is ( gently ) diverging near the limits p = 0 and p = 1 , but all the expectation values that we are interested in are finite . Our discussion here follows [10] , [12] very closely . Consider a set of possible neural responses labeled by i = 1 , 2 , … , K . The probability distribution of these responses , which we don't know , is given by p ≡ {pi} . A well studied family of priors on this distribution is the Dirichlet prior , parameterized by β , ( 7 ) Maximum likelihood estimation , which identifies probabilities with frequencies of occurrence , is obtained in the limit β → 0 , while β = 1 is the natural “uniform” prior . When K becomes large , almost any p chosen out of this distribution has an entropy very close to the mean value , ( 8 ) where ψ0 ( x ) = dlog2Γ ( x ) /dx , and Γ ( x ) is the gamma function . We therefore construct a prior that is approximately flat on the entropy itself by a continuous superposition of Dirichlet priors , ( 9 ) and we then use this prior to perform standard Bayesian inference . In particular , if we observe each alternative i to occur ni times in our experiment , then ( 10 ) and hence by Bayes' rule ( 11 ) Once we normalize this distribution we can integrate over all p to give the mean and the variance of the entropy given our data {ni} . In fact , all the integrals can be done analytically except for the integral over β [10] , [45] . Software implementation of this approach is available from http://nsb-entropy . sourceforge . net/ . This basic strategy can be supplemented in cases where we have prior knowledge about the entropies . In particular , when we are trying to estimate entropy in “words” of increasing duration T , we know that S ( T* , τ ) ≤S ( T , τ ) ≤S ( T* , τ ) +S ( T-T* , τ ) for any T*<T , and thus it makes sense to constrain the priors at T using the results from smaller windows T' , although this is not critical to our results . We obtain results at all integer values of T/τ for which our estimation procedure is stable ( see below ) and use cubic splines to interpolate to non-integer values as needed . There are two critical challenges to estimating the entropy of neural responses to natural signals . First , the overall distribution of ( long ) words has a Zipf-like structure ( Figure 4B ) , which is troublesome for most estimation strategies and leads to biases dependent on sample size . Second , the long correlation times in the stimulus mean that successive words ‘spoken’ by the neuron are strongly correlated , and hence it is impossible to guarantee that we have independent samples , as assumed implicitly in Eq . ( 10 ) . We tamed the long tails in the probability distribution by partitioning the space of responses , estimating entropies within each partition , and then using the additivity of the entropy to estimate the total . We investigated a variety of different partitions , including ( a ) no spikes vs . all other words , ( b ) no spikes , all words with one spike , all words with two spikes , etc . , ( c ) no spikes , all words with frequencies of over 1000 , and all other words . Further , for each partitioning , we followed [4] and evaluated S ( T , τ ) for data sets of different sizes αN , 0<α≤1 . By choosing fractions of the data in different ways we separated the problems of correlation and sample size . That is , to check that our estimates were stable as a function of sample size , we chose contiguous segments of experiment , while to check for the impact of correlations we ‘diluted’ our sampling so that there were longer and longer intervals between words . Obviously there are limits to this exploration ( one cannot access large , very dilute samples ) , but as far as we could explore the impact of correlations on our estimates was negligible once the samples sizes were sufficiently large . For the effects of sample size we looked for behavior of the form S ( α ) = S∞+S1/α+S2/α2 and took S∞ as our estimate of S ( T , τ ) , as in [4] . For all partitions in which the most common word ( silence ) was separated from the rest , these extrapolated estimates agreed and indicated negligible biases at all combinations of τ and T for which the 1/α2 term was negligible ( that is , did not change the extrapolation results by more than the extrapolation error ) compared to the 1/α; this happened for all τ≥0 . 5 ms at T≤25 ms . For smaller τ , estimation failed at progressively smaller T , and to obtain an entropy rate for large T we extrapolated to τ/T→0 using ( 12 ) where s ( τ ) was our best estimate of the entropy rate at resolution τ . All fits were of high quality , and the resulting error bars on the total entropy were negligible compared to those for the noise entropy . In principle , we could be missing features of the code which would appear only at high resolution for very long words , but this unlikely scenario is almost impossible to exclude by any means . Putting error bars on the noise entropy averaged over time is more difficult because these should include a contribution from the fact that our finite sample over time is only an approximation to the true average over the underlying distribution of stimuli . Specifically , the entropies were very different in epochs that have net positive or negative velocities . We constructed the repeated stimulus , v ( t ) = −v ( t+T0 ) , with T0 = 2 . 5 s . As a result , the sum Sn ( T , τ|t ) +Sn ( T , τ|t+T1 ) with T1≈T0 fluctuated much less as a function of t than the entropy in an individual slice . Because our stimulus had zero mean , every slice had a partner under this shift , and the small difference between T0 and T1 took account of the difference in latency between responses to positive and negative inputs . A plot of Sn ( T , τ|t ) +Sn ( T , τ|t+T1 ) vs . time t had clear dips at times corresponding to zero crossings of the stimulus , and we partitioned the data at these points . We derived error bars on the mean noise entropy 〈Sn ( T , τ|t ) t〉 by a bootstrap-like method , in which we constructed samples by randomly sampling with replacements from among these blocks , jittering the individual entropies Sn ( T , τ|t ) by the errors that emerge from the Bayesian analysis of individual slices . These blocks are long enough to preserve temporal correlations within them , but correlations across the block boundaries are negligible in the original signal , validating the procedure . As with the total entropy , we extrapolated to otherwise inaccessible combinations of T and τ , now writing ( 13 ) and fitting by weighted regression . Note that results at different T but the same value of τ were strongly correlated , and so the computation of χ2 was done using the full ( non-diagonal ) covariance matrix . The periodic term was important at small τ , where we could see structure as the window size T crossed integer multiples of the average interspike interval , τ0 = 2 . 53 ms . Error estimates emerged from the regression in the standard way , and all fits had χ2∼1 per degree of freedom . The procedures followed to get the total and noise entropy estimates in combination with the checks described above result in bias errors that are believed to be smaller than the random errors over the parameter range that we consider in all the analyses presented in this paper . Since there were no responses to repeated and unrepeated stimuli recorded at exactly the same illuminations , we used the data from the repeated experiment to evaluate both the noise entropy and the total entropy . We were looking for minute effects , so we tightened our analysis by discarding the first two trials , which were significantly different from all the rest ( presumably because adaptation was not complete ) , as well as excluding the epochs in which the stimulus was padded with zeroes . The remaining 98 trials were split into two groups of 49 trials each with the highest and the lowest ambient light levels . We then estimated the total entropy S ( h , l ) ( T , τ ) for the high ( h ) and low ( l ) intensity groups of trials , and similarly for the noise entropy in each slice at time t , . As above , assigning error bars was clearer once we formed quantities that were balanced across positive and negative velocities , and we did this directly for the difference in noise entropies , ( 14 ) where we allowed for a small difference in latencies between the groups of trials at different intensities . We found that ΔSn ( T , τ;t ) had a unimodal distribution and a correlation time of ∼1 . 4 ms , which allowed for an easy evaluation of the estimation error . | Neurons communicate by means of stereotyped pulses , called action potentials or spikes , and a central issue in systems neuroscience is to understand this neural coding . Here we study how sensory information is encoded in sequences of spikes , using a combination of novel theoretical and experimental techniques . With motion detection in the blowfly as a model system , we perform experiments in an environment maximally similar to the natural one . We report a number of unexpected , striking observations about the structure of the neural code in this system: First , the timing of spikes is important with a precision roughly two orders of magnitude greater than the temporal dynamics of the stimulus . Second , the fly goes a long way to utilize the redundancy in the stimulus in order to optimize the neural code and encode more refined features than would be possible otherwise . This implies that the neural code , even in low-level vision , may be significantly context dependent . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"mathematics/statistics",
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems"
] | 2008 | Neural Coding of Natural Stimuli: Information at Sub-Millisecond Resolution |
Translation of mRNA sequences into proteins typically starts at an AUG triplet . In rare cases , translation may also start at alternative non–AUG codons located in the annotated 5’ UTR which leads to an increased regulatory complexity . Since ribosome profiling detects translational start sites at the nucleotide level , the properties of these start sites can then be used for the statistical evaluation of functional open reading frames . We developed a linear regression approach to predict in–frame and out–of–frame translational start sites within the 5’ UTR from mRNA sequence information together with their translation initiation confidence . Predicted start codons comprise AUG as well as near–cognate codons . The underlying datasets are based on published translational start sites for human HEK293 and mouse embryonic stem cells that were derived by the original authors from ribosome profiling data . The average prediction accuracy of true vs . false start sites for HEK293 cells was 80% . When applied to mouse mRNA sequences , the same model predicted translation initiation sites observed in mouse ES cells with an accuracy of 76% . Moreover , we illustrate the effect of in silico mutations in the flanking sequence context of a start site on the predicted initiation confidence . Our new webservice PreTIS visualizes alternative start sites and their respective ORFs and predicts their ability to initiate translation . Solely , the mRNA sequence is required as input . PreTIS is accessible at http://service . bioinformatik . uni-saarland . de/pretis .
Translational initiation is a more complex process than reported in common textbooks . Experimental work showed that the canonical AUG–Methionine translational start is not always used to initiate eukaryotic translation [1–5] . Further alternative codons located upstream of the annotated AUG start can also serve as additional functional start sites and form additional or alternative ORFs . Those non–AUG triplets are postulated to differ from AUG by one nucleotide and hence comprise CUG , UUG , GUG , AAG , ACG , AGG , AUA , AUC and AUU [6] . Translation can proceed in–frame as well as out–of–frame relative to the main open reading frame [7] . This can , for example , lead to ( small ) upstream ORFs resulting in short peptides or to extended proteins when translational initiation takes place at an in–frame start codon located upstream of the canonical start . Ribosome profiling data provides information on the density of ribosomes located at different regions of the transcript upon application of small chemicals that block the elongation process [4 , 8] . Regions which are protected by ribosomes are not digested in the next step when the mRNA is treated with nucleases [9] . These ribosome footprints ( RNA ) have a length of about 30 nucleotides and are sequenced after nuclease treatment and subsequently mapped to a reference genome [9] . For example , Lee et al . ( 2012 ) applied ribosome profiling to human embryonic kidney 293 ( HEK293 ) cells [3] . As translational inhibitors they used Cycloheximide ( CHX ) and Lactimidomycin ( LTM ) , which both bind to the ribosome E-site [3] . While CHX can bind to both , initiating and elongating ribosomes , LTM prefers initiating ribosomes with a tRNA-empty E-site [3] . Thus , by combining both inhibitors , it is possible to differentiate initiating from elongating ribosomes [3] . Lee et al . identified 16 , 863 potential start sites out of about 10 , 000 transcripts whereby start sites were allowed to be located in the 5’ UTR , at the annotated start site , in the coding region , or in the 3’ UTR , respectively . Possible biological reasons underlying alternative translation initiation are the expansion of biological variety , regulatory processes as well as targeting of the proteins to different compartments [10–12] . Touriol et al . ( 2003 ) proposed that alternative translation initiation results in different proteoforms that can exhibit different functions as well as various cell localizations , which is of great importance for cell fate [12] . Some codons ( e . g . AUG or CUG ) are more frequently used as translation initiation starts than other codons [3 , 4] . So far , several bioinformatics studies have addressed the task of predicting alternative translational start sites or ORFs . The majority of these studies only considered AUG starts . Hatzigeorgiou ( 2002 ) applied an artificial neural network ( ANN ) embedding a linear search for AUG starts [13] . They achieved 94% accuracy and were able to predict the correct start site in 60% of human cDNAs . Saeys et al . ( 2007 ) developed a meta–tool that combines three simple AUG start site predictors that consider either position–weight–matrices , k–mer frequencies or the number of stop codons downstream of a start site . This combination of several simple predictors , named StartScan , resulted in a sensitivity of 80% , tested on human chromosome 21 [14] . Sparks and Brendel ( 2008 ) argued that when one only searches for one translational start , predicting the leftmost ( i . e . the most upstream ) AUG as sole correct translational start yielded specificity and sensitivity of 94% , respectively [15] . Chen et al . ( 2014 ) used a flexible window and represented human DNA as k–tupels that reflect the nucleotide composition and also integrated the physicochemical properties of amino acids [16] . For AUG codons , their method achieved an accuracy of 98% . A webservice of their algorithm is available . Besides , there also exist several web–based tools for ORF identification . ORF Finder searches for ORFs given the accession number or sequence and the genetic code [17] . ORF–Predictor provides an ab initio prediction of ORFs based on expressed sequence tag ( EST ) or cDNA sequences and BLASTX alignments or intrinsic sequence signals [18] . Only few studies involved ribosome profiling data or considered in– and out–of–frame start codons differing from AUG . Ivanov et al . ( 2011 ) studied annotated human 5’ UTRs via sequence alignments with orthologous species followed by a manual evaluation [6] to detect non-AUG initiation in human sequences . They predicted 42 novel genes with non–AUG upstream translation initiation . For 25 of these genes non–canonical translation initiation could be experimentally validated using Western blot as well as ribosome profiling data . They also confirmed 17 alternatively translated genes that were known at this time . Crappé et al . ( 2013 ) applied an SVM approach to ribosome profiling data to detect small conserved open reading frames ( sORFs ) in mouse that code for micropeptides ( 10 − 100 amino acids ) [19] . Baranov and colleagues ( 2014 ) used ribosome profiling data to calculate translational initiation probabilities [20] . In contrast to our work , they focused on the initiation strength of a putative start site as a function of the number of ribosome footprints . To our best knowledge , no study so far has evaluated the general properties of human start codons considering both AUG and all near–cognate codons , in- and out–of–frame , based on start sites identified by applying ribosome profiling , and exploited this to predict the initiation confidence from the mRNA sequence . The aim of this work was to analyze alternative translational start sites ( AUG and near–cognate codons ) with respect to sequence–based features to differentiate between true and false start sites . We used start sites that were identified by applying ribosome profiling to HEK293 cells [3 , 5] and mouse embryonic stem cells [4] as our primary datasets . Based on mRNA sequence information we generated support vector machines as well as a linear regression model for human and mouse sequences . The learned model can then be applied to mRNA sequences not covered by ribosome profiling data or to investigate the impact of mutations in the flanking sequence context of a start site on its translation initiation confidence . Our webservice PreTIS visualizes putative alternative start sites and the predicted initiation confidence in human .
Annotated genomic mRNA sequences for human and mouse were retrieved from Ensembl biomart ( Ensembl version 77 [21] ) . We only included curated mRNA sequences with available mRNA RefSeq identifier ( starting with NM_ ) . It was recently shown that 85% of the start sites used to initiate translation are conserved between human and mouse [3] . Thus , we used homologous pairs of human and murine sequences to calculate the conservation of putative start codons as well as the 5’ UTR sequence conservation ( see below ) . We identified the respective murine orthologous mRNA sequences using the approach by Ivanov et al . ( 2011 ) and used the first blastn [22] hit as the respective ortholog ( default blastn parameters ) . To identify putative start sites , each 5’ UTR was scanned for all AUGs and for alternative near–cognate start codons that differ from generic AUG by one nucleotide ( CUG , UUG , GUG , AAG , ACG , AGG , AUA , AUC und AUU ) and that are located either in–frame or out–of–frame with the main open reading frame . Different sequence–based features were then calculated for all putative start codons that have a downstream in–frame stop codon . To establish reliable true positive and true negative translational start site datasets for training and testing purposes , we used the findings from different ribosome profiling experiments [3–5] . Each dataset was analyzed independently . Note that the datasets used here contain translational start sites derived from ribosome profiling data by the original authors ( gene accession number , position relative to annotated start site , codon ) . We did not include raw ribosome profiling ( footprint ) data in our approach . In total , we trained two start site prediction models: a human prediction model based on the HEK293 dataset [3] and a mouse prediction model based on the Mouse ES dataset [4] . The third HEK293–AUG dataset [5] was used as validation set to further evaluate the reliability and robustness of the developed prediction model . For training and testing of every classifier , we considered each start site ( AUG and near-cognate ) that matched a start codon found by ribosome profiling as a true start . False start sites were defined as follows: remaining candidate start sites ( AUG and near-cognate ) that were not detected by ribosome profiling and that are , based on the assumption of a linear scanning model , located at least 99 nts downstream of the transcription start site as well as upstream of the most downstream reported true translation initiation start . Fig 1 shows an example mRNA sequence that illustrates the grouping of true positive and true negative start sites for training and testing purposes based on ribosome profiling data . This start sites categorization was executed for each of the three datasets , each time based on the individual ribosome profiling experimental results [3–5] . All features used here are solely based on information derived from the mRNA sequences . The prediction approach , shown in Fig 2 , was applied to the human HEK293 [3] and mouse ES datasets [4] . First , as mentioned , all putative start sites in the 5’ UTR were defined as true positives or true negatives based on the reported ribosome profiling data and their location in the mRNA sequence . We then balanced the size of the dataset so that it contains the same number of true and false start sites by randomly under–sampling from the larger dataset . We repeated the data balancing as well as the assignment of random training and test sets 10 times to evaluate the model robustness and reported the average performance . We applied Wilcoxon–rank sum test and Bonferroni correction ( with a significance threshold of 0 . 01 1 , 252 = 8 × 10 - 6 , with the total number of features as the denominator ) to test for the statistical significance of the biological , the k–mer , and the PWM features to differentiate between true and false start sites . We subsequently calculated all pairwise Pearson correlations between the significant biological and PWM features as well as for the 50 most significant k–mer features and only used uncorrelated ( |r| < 0 . 7 ) features in the training step . If two or more features were correlated , the one with the smallest p–value was used . The PWMs were calculated in each training step iteration to guarantee that the test set is independent on the calculated PWMs . All features were normalized ( mean zero and unit variance ) to ensure comparability . Next , we generated simple linear as well as support vector ( SVR ) regression models on 70% of this data and tested them on the remaining 30% of the data , using three different kernels for the SVM approach: linear , radial basis function ( RBF ) and polynomial . We applied 10–fold cross–validation to find the best penalty parameter C in [0 . 1 , 1 , 10 , 100] and ϵ-tube parameter in [0 . 01 , 0 . 1 , 1 , 10] for the training data set when applying SVR models . The remainder of parameters were kept at default values . Since we applied a regression approach , we subsequently applied 100 classification thresholds 0 . 0 ≤ t ≤ 1 . 0 in steps of 0 . 01 to the predicted output outpred in order to classify every start site as true or false based on its model outcome and the given threshold . These thresholds can be interpreted as initiation confidence where start sites with a regression value outpred ≥ t are predicted as true start sites and the ones with outpred < t as false start sites . If a start site is predicted with an initiation confidence > 1 , we substituted this value by 1 . The same holds for start sites with a predicted confidence < 0 , which were substituted with zero . We then compared the predicted class with the correct class and used the common measures accuracy , specificity , sensitivity , precision and area under the curve ( AUC ) as metrics for model assessment . The final model for the prediction of new mRNA sequences and for a SNP analysis was subsequently determined by comparing different model performances . The implementation was done in Python ( version 2 . 7 ) and using the Scikit–learn package ( version 0 . 17 ) for the machine learning part [34] . To investigate the effect of putative single nucleotide polymorphisms ( SNPs ) within the flanking sequence context of the start sites ( position -15 to +10 ) , we substituted ( in silico ) one nucleotide position at a time by all 3 remaining nucleotides , yielding 75 different contexts ( the start codon itself was not mutated ) . We then recalculated the needed sequence features to investigate the mutational impact and subsequently applied our final prediction model to all contexts . We then report the effect of these substitutions on the predicted initiation probabilities .
The start sites reported by ref . [3–5] based on ribosome profiling data were filtered to include only starts matching AUG and near–cognate codons in the 5’ UTR . For HEK293 cells [3] , this yielded 4 , 482 true start sites ( i . e . reported in the experimental analysis ) and 49 , 520 false start sites in 3 , 566 mRNAs . For mouse ES cells [4] , this gave 3 , 009 true start sites and 19 , 864 false start sites in 1 , 632 mRNAs . True ( reported ) starts were assumed to be true positives ( TP ) and false ( not reported and upstream of the most downstream reported ) starts were assumed to be true negatives ( TN ) . For comparison , we also included a smaller dataset of Ohler and colleagues [5] who only determined AUG starts in HEK293 cells . Table 1 displays the three datasets . All reported analyses are based on these filtered datasets . Among the considered AUG and near-cognate start codons , AUG ( human: 26% , mouse: 16% ) , CUG ( human: 30% , mouse: 34% ) and GUG ( human: 13% , mouse: 19% ) were the most prevalent translational start codons , see S1 Fig . Thus , CUG and GUG are more often used in mouse compared to human . This is in accordance with [3 , 4] and shows that the start codon itself is very important for translational initiation . Table 1 illustrates that the negative sets outnumber the positive sets by factors of 7 ( mouse ES ) and 11 ( HEK293 ) . To avoid a class size dependent bias , we randomly under–sampled the same number as true positive start sites from the true negative set . Next , we trained on 70% and tested on 30% of the data ( randomly assigned ) . Table 2 lists the performance of human and mouse models together with the optimal thresholds t . All human models perform very similarly with accuracies of about 80% while the average performance of the mouse model is lower with average accuracies of about 76% , see Table 2 . We also computed receiver operating characteristic ( ROC ) curves and the associated area under the curve ( AUC ) . In accordance with the other metrics , also the AUC values were satisfactory with average values of about 80% and 76% for the human and mouse models , respectively ( Table 2 ) . To investigate the transferability of our best human prediction model , we analyzed its performance using the mouse ES data as well as the HEK293–AUG dataset , see Table 4 . With the threshold of t = 0 . 54 that was found to be optimal for the trained HEK293 dataset , we obtained for the mouse ES dataset an accuracy of 76% , a sensitivity of 72% and a specificity of 77% . By scanning all possible thresholds , see S4 ( A ) Fig , we found that t = 0 . 52 yields a more balanced performance of 75% , 76% and 74% for accuracy , sensitivity and specificity , respectively . Decreasing the threshold seems to be advantageous for the mouse data set , since some true positives seem to possess weaker features for translational initiation ( e . g . a weak flanking sequence context or a less common initiation codon ) , but are nevertheless true positive starts . We then applied our best regression model to the start sites reported in the HEK293–AUG dataset that only contains AUG starts [5] . The categorization of true positive and true negative start sites was conducted as above for the HEK293 dataset ( see Fig 1 ) , with the only difference that the HEK293–AUG dataset only contains AUG start sites instead of AUG and all near-cognate codons . Thus , we defined again the false start sites as all AUG starts located in the 5’ UTR that were not detected by ribosome profiling and are located upstream of the most downstream true start site . Differentiating only between true and false AUG start sites is particularly difficult because the AUG itself is a very strong signal for a true start site and just by random chance there might be AUGs with , for example , good flanking sequence , which are not used as translational start sites ( or are not reported in the dataset ) . Moreover , our prediction model was trained on all possible cognate codons instead of AUG alone . Our best model with the determined threshold of t = 0 . 54 detected 77% of the true AUG starts in the HEK293–AUG dataset ( sensitivity of 77% ) . Nevertheless , the specificity of this prediction is only 44% and thus the overall accuracy is only slightly better than a random decision ( 58% ) , compare to Table 4 . However , when increasing the threshold from t = 0 . 54 to t = 0 . 65 , we were able to increase the overall accuracy to 63% . A threshold of t = 0 . 65 was found to be optimal for this dataset , see S5 ( A ) Fig . Problematic was here the precision ( i . e . the number of true positives out of all samples classified as positive ( T P s T P s + F P s ) ) . Many starts that we assumed to be true negatives actually show properties of true positives and are therefore classified as false positives . Especially if the dataset is highly unbalanced ( e . g . the number of mouse ES true starts is only 15% of the false start sites ) this effect has a strong influence on the precision . When we balanced our datasets , the precision increased drastically from 31% to 75% for the mouse ES dataset and t = 0 . 52 and from 56% to 63% for the HEK293–AUG dataset and t = 0 . 64 , see Table 4 . The established prediction model can , for example , be used to predict translational start sites which are not covered by ribosome profiling experiments or to analyze the impact of mutations in the flanking sequence around the start site .
PreTIS is a webservice to predict the initiation confidence of all reading–frame independent start sites ( AUG and all near-cognate codons ) located in the 5’ UTR of human mRNA sequences . Thereby , the best human prediction model described here is used as underlying regression model . The webservice application PreTIS requires the mRNA sequence and is accessible at http://service . bioinformatik . uni-saarland . de/pretis . Based on the given human mRNA sequence , all possible AUG and near-cognate start sites , with a surrounding window of at least ±99 nts ( needed to calculate k-mers ) and an in–frame downstream stop codon , are located in the 5’ UTR . PreTIS then calculates the required sequence features ( see Table 3 ) for all detected start sites and subsequently predicts the initiation confidence . Based on the predicted initiation confidence value and the given prediction threshold of t = 0 . 54 , a start site is categorized into different initiation confidence classes . For start sites with confidence values c greater than the given threshold t , the four confidence groups were defined as follows: very high ( hot/best candidates with c ≥ 0 . 9 ) , high ( 0 . 8 ≤ c < 0 . 9 ) , moderate ( 0 . 7 ≤ c < 0 . 8 ) and low ( t ≤ c < 0 . 7 ) confidence , respectively . Especially start sites with very high confidence values can be considered as hot candidates for translational initiation . The predicted initiation confidence for each start site is visualized by barplots with the x-axis displaying the mRNA position ( compare to Fig 5 ) . This enables a comprehensive comparison of , for example , different flanking sequence contexts . Features calculated for each start site can also be downloaded as . csv file for further analyses . For the calculation of some features , an orthologous mouse sequence is required . This is automatically implemented by the embedded blastn [22] search . The mRNA sequence found by blastn can be inspected afterwards and replaced , if desired . Furthermore , each job is given a Session– and a Job–ID , which enables unambiguous accession to the prediction results . Thus , PreTIS is an intuitive tool that solely requires the human mRNA sequence as input . It gives access to various calculated sequence-encoded and experimentally shown important sequence properties for translational initiation . In addition , an initiation confidence value for each start site is calculated using an established regression model that is based on recent experimental data . AUG as well as alternative start codons—in and out of the main reading frame—are considered . | Ribosome profiling data and mRNA sequence features can be used to build reliable classification models with accuracies of about 80% for start codon and open reading frame prediction in human . All predicted start sites of one transcript are postulated to have the potential to initiate translation . They could , for example , be used in different tissues or in a specific cellular condition , such as stress response . Although there exist already several other approaches to predict translational initiation start sites , so far none of them considers all in– and out–of–frame AUG and near–cognate codons . The provided web service PreTIS considerably simplifies and assists the analysis of mRNA sequences in terms of prediction of possible translation start sites and their visualization . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"messenger",
"rna",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"untranslated",
"regions",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"translation",
"initiation",
"sequence",
"analysis",
"sequence",
"alignment",
"mathematical",
"and",
"statistical",
"techniques",
"gene",
"expression",
"molecular",
"biology",
"ribosomes",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"protein",
"translation",
"genetics",
"5'",
"utr",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods"
] | 2016 | PreTIS: A Tool to Predict Non-canonical 5’ UTR Translational Initiation Sites in Human and Mouse |
Clonorchis sinensis infection elicits hepatic inflammation , which can lead to cholangitis , periductal hepatic fibrosis , liver cirrhosis , and even cholangiocarcinoma . Hepatic macrophages are an intrinsic element of both innate and acquired immunity . This study was conducted to demonstrate the dynamics of hepatic macrophage polarization during C . sinensis infection in mice and to identify factors regulating this polarization . Treatment of hepatic macrophages isolated from normal mice with C . sinensis excretory/secretory products ( ESPs ) resulted in the preferential generation of classically activated hepatic macrophages ( M1 macrophages ) and the production of pro-inflammatory cytokines . Additionally , cells stimulated with C . sinensis ESPs exhibited changes in cellular morphology . During the early stages of C . sinensis infection , hepatic macrophages preferentially differentiated into M1 macrophages; however , during the C . sinensis mature worm stage , when eggs are released , there were significant increases in the abundance of both M1 macrophages and alternatively activated hepatic macrophages ( M2 macrophages ) . Moreover , there was a further increase in the M2 macrophage count during the fibrotic and cirrhotic stage of infection . Notably , this fibrotic and cirrhotic stage promoted a strong increase in the proportion of Arg-1-producing macrophages ( M2 phenotype ) , which were associated with fibrosis and tissue repair in the liver . Our results suggest that the dynamic polarization of hepatic macrophages as C . sinensis infection progresses is related to the histological lesions present in liver tissue . Hepatic macrophages thus play an important role in local immunity during C . sinensis infection .
Clonorchis sinensis infection represents a major public health problem in Asia , with approximately 30 million people infected , of which 1 . 3 million reside in Korea [1 , 2] . C . sinensis infection causes several pathological alterations in the bile ducts , including hyperplasia of the mucosa , dilatation of the bile duct , and fibrosis in the region surrounding the bile duct . Chronic and heavy infections can cause cholangitis or abscess formation , pancreatitis , gallstone formation , and cholangiocarcinoma [1–3] . C . sinensis infection is distinguished by an adaptive immune response related to a Th2 phenotype [4] . Although Th2-type cytokines and other host-protective immune responses attack parasites , the parasites are in turn able to control the host immune response to maintain survival , consequently leading to chronic infection [5] . However , despite the presence of C . sinensis in the bile duct , only few reports have been published regarding the local immune response to the parasite within the liver and bile duct . Macrophages can be classified into two subpopulations: classically activated macrophages ( M1 macrophages ) and alternatively activated macrophages ( M2 macrophages ) . M1 macrophages are distinguished by the expression of high levels of inducible nitric oxide synthase ( iNOS ) , tumor necrosis factor ( TNF ) -α , CD16/32 , and chemokines such as chemokine ( C-X-C ) motif ligand ( CXCL ) 9 , CXCL10 , and CXCL11; these macrophages are generated in response to pro-inflammatory stimuli such as lipopolysaccharide ( LPS ) [6–10] . Conversely , M2 macrophages are distinguished by their high expression of arginase-1 ( Arg-1 ) , interleukin ( IL ) -10 , CD206 , and chemokines such as C-C motif chemokine ligand ( CCL ) 2 and CCL22 , and develop in response to inflammatory stimuli such as IL-4 [11–13] . However , the specific contributions of hepatic macrophage proliferation to infection control and/or the restoration of tissue homeostasis remain unclear . Macrophages differentiate into M1 and M2 functionalities following infection with microorganisms or parasites [10] . In this regard , the complex interaction between excretory/secretory proteins ( ESPs ) from Fasciola hepatica and peritoneal macrophages is crucial for the establishment of this parasite in mice . Moreover , treatment with F . hepatica ESPs has been shown to induce alternatively activated macrophages in experimental models [11 , 13] . Among hepatic macrophages , resident macrophages termed Kupffer cells evoke an important element of innate immunity and display an early , rapid response to dangerous stimuli [9 , 14] . Another group of hepatic macrophages appears to be derived from circulating monocytes . In particular , bone marrow-derived monocytes migrate into various organs , where they are converted into organ-specific macrophages [15–17] . Upon activation , these different classes of hepatic macrophages release products such as cytokines , chemokines , nitric oxide ( NO ) , and reactive oxygen species ( ROS ) . Thus , hepatic macrophages are involved in the liver’s immune response to infection and other stressors [18 , 19] . To the best of our knowledge , the dynamics of hepatic macrophage polarization during C . sinensis infection have not been reported . In this study , we analyzed M1 and M2 phenotypes to identify the presence and degree of macrophage polarization in the liver upon infection with C . sinensis . Possible roles for C . sinensis ESPs in hepatic macrophage polarization were also investigated . These results may provide insights into the development of methods to better control hepatic fibrosis in clonorchiasis .
The animal experiment protocol was approved and reviewed by the Institutional Animal Care and Use Committee ( IACUC ) of Yonsei University Health System , Seoul , Korea ( approval no . 2015–0339 ) and followed the National Institutes of Health ( NIH ) guidelines for the care and use of laboratory animals ( NIH publication no . 85–23 , 1985 , revised 1996 ) . The animal facility was certified by the Ministry of Food and Drug Administration and by the Ministry of Education , Science , and Technology ( LNL08-402 ) . Animal experiments were carried out in animal biosafety level-3 ( ABL-3 ) facilities in accordance with ABL-3 standard managing practices . C . sinensis metacercariae were obtained from the freshwater fish Pseudorasbora parva , caught at Sancheong , Gyeongsangbuk-do , Korea . Male , 5-week-old BALB/c mice were purchased from Orientbio ( Seongnam , Korea ) . The animals were randomly divided into uninfected or infected groups ( n = 10 for each group ) . Mice in the infected group were orally infected with 30 metacercariae . The experiment was designed to focus on the juvenile worm and egg-emission stages of infection . As such , the first group of mice was studied during the early infection stage , 1 to 2 weeks post-infection . The second and third groups were studied 7 and 8 weeks post-infection , respectively , representing the egg-production stage of mature worms , and the fourth group was studied 10 , 12 , and 16 weeks post-infection , during the fibrotic and cirrhotic stage . Mice were euthanized under deep anesthesia with ethyl ether and each experiment was performed in duplicate . Male , 5-week-old Sprague-Dawley ( SD ) rats ( Orientbio Inc . ) were individually infected with 50 metacercariae of C . sinensis . At 8 weeks post-infection , the rats were euthanized under deep anesthesia with ethyl ether , before adult worms were isolated from the bile ducts , washed five times with phosphate-buffered saline ( PBS ) containing 100 U/ml penicillin/streptomycin ( Sigma Aldrich , St . Louis , MO , USA ) , then incubated for 24 h at 37°C with 5% CO2 . After incubation , the medium was centrifuged for 10 min at 1 , 000 × g to remove the worms and cellular debris . The supernatant was then centrifuged for a further 10 min at 18 , 000 × g before being filtered with a syringe-driven 0 . 45 μm filter . The concentration of protein was measured using Bradford assay reagent ( Thermo Fisher Scientific , Waltham , MA , USA ) . The livers of five mice from each group were removed , and five transverse sections were produced per lobe . The liver sections were then fixed in 10% phosphate-buffered neutral formalin . The specimens were embedded in paraffin and stained with hematoxylin and eosin ( H&E ) and periodic acid-Schiff ( PAS ) to observe any histopathological lesions . Nonparenchymal liver cells were isolated via the pronase-collagenase method , as described previously [18] . Briefly , portal veins were intubated with a plastic catheter ( 2 mm diameter ) , and livers were perfused in situ with 10 ml PBS containing 0 . 1% mixed type IV collagenase ( 10 ml/min ) at 37°C in a nonrecirculating design to remove red blood cells . Livers were then transferred to 35 mm culture dishes and minced into small pieces . The liver tissues were dispersed in 10 ml Roswell Park Memorial Institute 1640 medium ( RPMI 1640; Hyclone , Logan UT , USA ) containing 0 . 1% type IV collagenase at 37°C for 30 min , then mixed gently with a graduated pipette for 10 min . After digestion , the liver homogenates were filtered through a 74-μm stainless steel wire mesh to remove undigested tissue , and the resulting cell suspensions were centrifuged at 300 × g ( 5810R; Eppendorf , Hamburg , Germany ) for 5 min at 4°C . The aqueous phase was discarded , and the cell sediment was suspended in RPMI 1640 , transferred into a new 10-ml centrifuge tube , and centrifuged at 300 × g for 5 min at 4°C . Again , the aqueous phase was discarded , and the resulting cell sediment was seeded into a T-75 flask at a density of 1–3 × 108 cells/well in Dulbecco’s modified Eagle’s medium ( DMEM; Hyclone ) supplemented with 10% fetal bovine serum ( FBS; Hyclone ) and 100 U/ml penicillin/streptomycin ( Sigma-Aldrich ) . Cells were then incubated for 2 h in a 5%-CO2 atmosphere at 37°C . Nonadherent cells were removed from the dish; the adherent cells were considered hepatic macrophages . Hepatic macrophages were purified from normal mice and then cultured in medium containing either PBS , 10 μg/mL ESPs , 100 ng/mL LPS ( Sigma Aldrich ) , or 10 ng/ml IL-4 ( Peprotech Inc . , Rocky Hill , NJ , USA ) for 48 h at 37°C . The macrophages treated with LPS or IL-4 were regarded as positive controls for M1 and M2 macrophages , respectively . Following in vitro treatment , macrophages were stimulated with C . sinensis ESPs ( 10 μg/mL ) , and supernatants were harvested after 48 h . TNF-α , IL-6 , and IL-13 levels were determined by enzyme-linked immunosorbent assays ( ELISAs ) , according to the manufacturer’s instructions ( Peprotech Inc . ) . Single-cell suspensions were adjusted to 1 × 106 cells/100 μL PBS containing 1% FBS . To analyze hepatic macrophage purity , cells were incubated with a fluorescein isothiocyanate ( FITC ) -conjugated antibody specific for mouse F4/80 , a macrophage/microglial marker , and an APC-conjugated antibody specific for mouse CD11b ( BioLegend , San Diego , CA , USA ) . For M1 and M2 surface marker analysis , cells were incubated for 1 h at 4°C with either PE-conjugated or Alexa 647-conjugated antibodies specific for mouse CD16/32 or CD206 ( BioLegend; 10 μg/mL each ) , respectively , then washed with PBS . The hepatic macrophages were then fixed with 1% paraformaldehyde/PBS and analyzed using a BD FACSCalibur Flow Cytometer ( BD Biosciences , San Jose , CA , USA ) . Results were analyzed using FlowJo software ( BD Biosciences ) . For microscopic examination , the hepatic macrophages were washed three times with cold PBS , fixed with 2% paraformaldehyde in PBS for 30 min , then permeabilized with 0 . 2% ( w/v ) Triton X-100 in PBS for 5 min . Hepatic macrophages were then blocked with 0 . 5% bovine serum albumin in PBS for 1 h before being incubated with an Alexa-Fluor 488 Avidin-labeled antibody detecting F4/80 , a macrophage marker ( Invitrogen , Carlsbad , CA , USA ) . Nuclei were visualized by staining with 4' , 6-diamidino-2-phenylindole ( DAPI ) , and cells were observed under an LSM PASCAL confocal laser scanning microscope ( Carl Zeiss , Oberkochen , Germany ) . For immunohistochemistry analyses , heat-induced epitopes in liver sections were retrieved by incubation with proteinase K at 100°C in a water bath for 1 h , followed by incubation with an anti-rabbit F4/80 antibody ( 1:50 dilution; LifeSpan Biosciences , Seattle , WA , USA ) for 24 h at 4°C in a humidified chamber . After washing with PBS , the slides were incubated with an anti-mouse secondary antibody ( LifeSpan Biosciences ) for 1 h at room temperature , and visualized using diaminobenzidine and hematoxylin as the counter stain . After detection by F4/80 staining , hepatic macrophages were semi-quantified by viewing 10 fields at 100× magnification , using an Olympus light microscope ( Tokyo , Japan ) . Total RNA was isolated from hepatic macrophages , using TRIzol reagent ( Invitrogen ) and converted to cDNA , using a Reverse Transcription Kit according to the manufacturer’s instructions ( Fermantas Life Sciences , Waltham , MA , USA ) . Quantitative real-time PCR was then conducted using SYBR Green Master ( Rox ) reagents ( Roche Diagnostics , Basel , Switzerland ) and a 7300 Real-time PCR machine ( Applied Biosystems , CA , USA ) . The reaction conditions were as reported previously [20]: stage 1 , 50°C for 2 min; stage 2 , 95°C for 10 min; stage 3 , 45 cycles of 95°C for 15 s and 60°C for 1 min , followed by a melting curve analysis process . Fold changes in gene expression were calculated using the 2−ΔΔCt method . The sequences of the primer pairs used for these analyses were as follows: Tnfa ( forward ) 5′-CATCTTCTCAAAATTCGAGTGACAA-3′ and ( reverse ) 5′-TGGGAGTAGACAAGGTACAACCC-3′ [20]; Il10 ( forward ) 5′-ACTTTAAGGGTTACTTGGGTTGC-3′ and ( reverse ) 5′-ATTTTCACAGGGGAGAAATCG-3′; Cxcl9 ( forward ) 5′-TCTCGGACTTCACTCCAACACA-3′ and ( reverse ) 5′-ACTCCACACTGCTGGAGGAAGA-3′; Ccl2 ( forward ) 5′-AAGCCAGCTCTCTCTTCCTCCA-3′ and ( reverse ) 5′-AAGCCAGCTCTCTCTTCCTCCA-3′; Nos2 ( encoding the iNOS protein ) ( forward ) 5′-GCCACCAACAATGGCAACA-3′ and ( reverse ) 5′-CGTACCGGATGAGCTGTGAATT-3′; Arg1 ( forward ) 5′-CAGAAGAATGGAAGAGTCAG-3′ and ( reverse ) 5′-CAGATATGCAGGGAGTCACC-3′ [20]; and Gapdh ( forward ) 5′-GGTGAAGGTCGGTGTGAACG-3′ and ( reverse ) 5′-ACCATGTAGTTGAGGTCAATGAAGG-3′ . The values in each graph represent the means ± standard deviations ( SDs ) of the results obtained from independent experiments . Data were analyzed by Student’s t-tests , and p-values of less than 0 . 05 were considered statistically significant .
Compared with that in controls , ESP stimulation significantly increased the proportion of M1 macrophages but not M2 macrophages ( Fig 1A ) . Notably , the macrophages treated with 10 μg/ml C . sinensis ESPs exhibited both an increase in cell size and a “spiked” cell shape , characteristic of activated macrophages ( Fig 1B ) . Hepatic macrophages from uninfected mice produced TNF-α , IL-6 , and IL-13 in response to ESP stimulation . In particular , the stimulated cells secreted significantly more amount of the pro-inflammatory cytokines TNF-α and IL-6 than the control cell population ( Fig 2 ) . C . sinensis-infected mice showed obvious infiltration of inflammatory cells and fibrocystic accumulation around intrahepatic bile ducts ( Fig 3 ) . Increasing mucin deposits within the liver were observed along with the development of fibrosis and cirrhosis . The morphological changes that occurred during clonorchiasis in mice are shown in Fig 3 . Notably , hyperplasia of the bile duct epithelium was observed during early infection ( Fig 3 ) . At 1–2 weeks post-infection , there was a marked increase in the number of mucin-positive cells , mainly at the cholangiocytes . A juvenile worm arrowheads ) was observed in the bile ducts , along with moderate periductal fibrosis ( arrow ) and bile duct proliferation . Mucin staining was strongly positive at the chronic stage . An adult worm ( black arrowhead ) was observed in the dilated bile duct , and massive fibrosis ( arrow ) was visible around the bile duct ( Fig 3 ) . Additionally , acute inflammatory cell infiltration around the bile duct and portal areas , accompanied by cholangiocyte proliferation , was observed 2 weeks post-infection . At the egg-emission stage , one adult worm ( black arrowhead ) was apparent in a dilated bile duct , and considerable infiltration by inflammatory cells was observed in the surrounding bile ducts . Minor collagen deposition was observed , and intrahepatic bile duct epithelium was noted . Intrahepatic bile ducts gradually became more extended until 7–8 weeks post-infection ( Fig 3 ) . In contrast , inflammatory cell infiltration appeared to be attenuated inside the bile duct after the parasite was removed , whereas collagen deposition and mucin expression were observed around the bile ducts of the livers of infected mice ( Fig 3 ) . These results also indicated that there was a positive correlation between the repaired bile ducts and mucin secretion in cholangiocytes . Consistent with our in vitro analyses , hepatic macrophages isolated from mice infected with C . sinensis exhibited the characteristic spindle shape of activated macrophages ( Fig 4 ) . As early as 1 week postinfection , an increase in the relative proportion of hepatic M1 macrophages was observed within the livers of infected mice; the percentage of M1 macrophages in infected mice then increased steadily until the egg-production stage , but decreased sharply at the cirrhotic stage . In contrast , the percentage of M2 macrophages decreased sharply at the beginning of the egg production stage and then peaked during the chronic stage ( Fig 6 ) . As shown in Fig 7 , expression of the M1-specific enzyme iNOS began to increase 1 week after infection , and increased steadily until the egg-production stage . However , its expression was significantly reduced at the fibrotic and cirrhotic stage . Conversely , the expression of Arg-1 , an M2-specific enzyme , increased from 7 weeks after infection and was markedly upregulated 10–12 weeks postinfection . Interestingly , iNOS was highly upregulated at both the migration and egg-emission stages , whereas Arg-1 was strongly upregulated after the egg-migration stage . There was an increase in the expression of M1-related chemokines and CXCL9 at the peak of the egg-release stage . Additionally , a marked increase in the expression of the M2-related chemokine CCL2 was observed 1 week after infection , and expression of this factor was maintained at a high level until week 16 ( Fig 8 ) . In addition , the M1-related cytokine TNF-α was highly upregulated during acute infection , whereas IL-10 expression increased at the peak of the egg emission stage .
C . sinensis infection is known to primarily activate humoral immunity and Th2-type immune responses , but some studies have also described the activation of cellular immunity , particularly at the site of infection [1 , 4 , 21–24] . In this study , we describe for the first time , hepatic macrophage polarization in mice liver following C . sinensis infection , as well as the secreted factors that could be involved in the immune response . We show that the polarization of hepatic macrophages shifts dynamically during the different stages of C . sinensis infection . C . sinensis ESPs favored the generation of M1-type hepatic macrophages in vitro . In addition , during the early stage of infection , C . sinensis induced morphological changes in murine hepatic macrophages , which were functionally polarized into M1 . On the other hand , M2 hepatic macrophages were increased in number during the egg-emission stage of infection , with this increase accelerating during the fibrotic and cirrhotic stage of infection . As components of the host immune system , macrophages regulate tissue generation , help to maintain organ function , regulate metabolic mechanisms , and promote inflammatory responses [19] . This range of responses is possible because macrophage activation is dynamic , and they are able to change and adapt in response to changes in the local environment [18 , 20] . Resident hepatic macrophages , known as Kupffer cells , account for 80–90% all of tissue macrophages throughout the body [14] . Notably , C . sinensis ESP secretion results from the direct interaction between C . sinensis antigens and host cholangiocytes in the bile duct . Such interactions between C . sinensis and host cells , including hepatic macrophages , drive specific immune responses in the liver , demonstrating the importance of macrophages in host defense mechanisms . Infection with the parasite T . spiralis reportedly induces a shift towards proliferative M1 macrophages during the early stages , with M2 macrophages becoming more prevalent once the parasite progresses to the larval stage and forms cysts in muscle tissue [25] . Several studies have suggested that M1 macrophages are capable of killing schistosomula by producing nitric oxide [19–20] , which plays a role in preventing the progression of fibrosis . Although our results show that M1-related factors , such as iNOS , TNF-α , and CXCL9 , were observed when the M1 ratio peaked during the egg-emission stage , we were unable to determine whether adult worms were expelled from the bile ducts . In contrast , M2 macrophages are thought to contribute to schistosome-induced fibrosis via the metabolism of Arg-1 [18–20] . Indeed , M2-specific enzymes , such as Arg-1 , and M2-related chemokine activation of the CCL2-CCR2 axis are associated with monocyte infiltration in patients with chronic liver disease and fibrogenesis [14 , 15] . In a mouse model of inflammatory bowel disease , peritoneal M2 macrophages induced by T . spiralis suppressed the production of pro-inflammatory cytokines such as TNF-α and IL-6 , and consequently reduced the severity of inflammation [25 , 26] . In the present study , Arg-1 levels were maintained throughout the infection , with the highest level occurring at the fibrotic and cirrhotic stage . A previous study showed that the worm recovery rate decreased with prolonged infection , and that worms were expelled from mice over time [23] . Another study reported that the number of Kupffer cells , acting as antigen presenting cells , is increased 70-fold in the early stages of clonorchiasis when compared to that in controls [24] . However , until now , there has been no convincing evidence demonstrating that M2 macrophages are involved in the expulsion of worms from mice . Our data showed very little evidence of the presence of worms in the bile ducts , using histological methods more than 10 weeks after infection , and the increase in the proportion of M2 macrophages , involved in tissue regeneration and the remodeling of the surrounding bile duct , at these stages was consistent with a strong immune response . Furthermore , we detected the expression of mucin at the site of worm expulsion . Mucin was previously shown to be overexpressed in the intestine during an immune response to hookworm infection and is known to contribute to the expulsion of hookworms [27] . Consistent with this , we observed convincing collagen deposition and mucin expression in the liver during the fibrotic and cirrhotic stage of C . sinensis infection . These results suggest a correlation between the repair of bile ducts following worm expulsion and mucin secretion in cholangiocytes , although the connection between hepatic macrophages and mucin secretion following C . sinensis infection is currently unclear . In future studies , it will be interesting to examine the relationships between the phenotypes of polarized hepatic macrophages and mucin-secreting cells . Furthermore , our results indicate that M2 hepatic macrophages modulate fibrosis and the tissue repair processes that are necessary to transform bile ducts from the thickened , dilated morphology caused by C . sinensis infection to a narrower morphology . Further studies are needed to determine whether immunomodulation depends on specific C . sinensis ESP molecules in vitro or whether the elimination of murine M2 macrophages in vivo affects worm expulsion or survival . In summary , our results suggest that the modulation of hepatic macrophage polarization by C . sinensis infection may serve as a potential mechanism for both parasite immune escape and host tissue repair . Furthermore , the dynamic polarization of hepatic macrophages as infection progresses corresponds to histological lesions within liver tissue . Hepatic macrophages thus play an important role in local immunity during C . sinensis infection . | Infection with Clonorchis sinensis is a major public health problem in Asia , resulting in loss of liver function and chronic liver diseases , including cancers . However , to the best of our knowledge , the immune response to fluke infection in the liver has not been systematically investigated . Here , we demonstrated that C . sinensis infection triggered a shift in the characteristics of macrophages , the primary cells associated with host immunity , throughout the course of the worm’s life cycle . Eventually , the increased number of M2 macrophages may result in fibrosis and the remodeling of bile ducts within the liver . Our findings suggest that , while the infection initially led to changes in the immune response that facilitated C . sinensis survival , ongoing infection may also reduce the severity of disruption of liver function . Hepatic macrophages activated during C . sinensis infection may not only be operating in histological lesions around the bile duct , but may also play a role in local immunity . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"biliary",
"system",
"invertebrates",
"cell",
"motility",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"liver",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"helminths",
"immunology",
"fibrosis",
"parasitic",
"diseases",
"animals",
"trematodes",
"clonorchis",
"sinensis",
"developmental",
"biology",
"molecular",
"development",
"white",
"blood",
"cells",
"animal",
"cells",
"flatworms",
"immune",
"response",
"chemotaxis",
"immune",
"system",
"bile",
"ducts",
"cell",
"biology",
"anatomy",
"clonorchis",
"physiology",
"chemokines",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2017 | Clonorchis sinensis antigens alter hepatic macrophage polarization in vitro and in vivo |
The host protein CPSF6 possesses a domain that can interact with the HIV-1 capsid ( CA ) protein . CPSF6 has been implicated in regulating HIV-1 nuclear entry . However , its functional significance for HIV-1 replication has yet to be firmly established . Here we provide evidence for two divergent functions of CPSF6 for HIV-1 replication in vivo . We demonstrate that endogenous CPSF6 exerts an inhibitory effect on naturally occurring HIV-1 variants in individuals carrying the HLA-B27 allele . Conversely , we find a strong selective pressure in these individuals to preserve CPSF6 binding , while escaping from the restrictive activity by CPSF6 . This active maintenance of CPSF6 binding during HIV-1 CA evolution in vivo contrasts with the in vitro viral evolution , which can reduce CPSF6 binding to evade from CPSF6-mediated restriction . Thus , these observations argue for a beneficial role of CPSF6 for HIV-1 in vivo . CPSF6-mediated restriction renders HIV-1 less dependent or independent from TNPO3 , RanBP2 and Nup153 , host factors implicated in HIV-1 nuclear entry . However , viral evolution that maintains CPSF6 binding in HLA-B27+ subjects invariably restores the ability to utilize these host factors , which may be the major selective pressure for CPSF6 binding in vivo . Our study uncovers two opposing CA-dependent functions of CPSF6 in HIV-1 replication in vivo; however , the benefit for binding CPSF6 appears to outweigh the cost , providing support for a vital function of CPSF6 during HIV-1 replication in vivo .
An essential part of the HIV-1 lifecycle is the transfer of its genetic material from the cytoplasm into the nucleus for subsequent integration into the host genome . In actively proliferating cells , breakdown of the nuclear membrane during mitosis ensures viral access to the host chromosomes . However , HIV-1 and other lentiviruses share the ability to efficiently infect non-dividing cells [1]–[3] . This necessitates a mechanism of hijacking the cellular transport machinery in order for HIV-1 to cross the intact nuclear envelope through nuclear pores [4] , [5] . Understanding the mechanism of HIV-1 nuclear entry is crucial [6] , [7] , as this is the property that enables HIV-1 to infect such critical target cell types in vivo as resting or partially activated CD4+ T cells [8] , [9] as well as tissue macrophages [10] . Comparative studies utilizing HIV-1 and murine leukemia virus ( MLV ) , a virus unable to efficiently infect non-dividing cells , demonstrated that the viral capsid ( CA ) protein is the major determinant for HIV-1 infection of non-dividing cells [11] . Since MLV is blocked in non-dividing cells [12] , [13] at nuclear entry [14] , HIV-1 must be equipped with a CA-dependent mechanism to utilize the host nuclear transport machinery to infect non-dividing cells . Indeed , there is mounting evidence pointing to the role of CA in HIV-1 nuclear entry [11] , [15]–[19] . Therefore , one major question is how CA orchestrates interactions between pre-integration complexes ( PICs ) and host cellular machinery to promote HIV-1 nuclear entry . Genome-wide siRNA screenings revealed a number of potential cellular factors that could affect HIV-1 nuclear import [20]–[22] . Among these host molecules , transportin 3 ( TNPO3 ) , RanBP2 and Nup153 are of particular interest for the following reasons: 1 ) Knockdown of these molecules blocks HIV-1 infection after reverse transcription but before integration [20] , [21] , [23] . 2 ) These molecules are required by HIV-1 but not by MLV [20] , [21] , [23] . 3 ) Most importantly , HIV-1 usage of these molecules is determined by CA [17]–[19] , [24]–[26] . These findings provide a CA-dependent link between HIV-1 infection of non-dividing cells and host factors exploited by HIV-1 to promote nuclear entry . An unresolved question is how CA mediates HIV-1 utilization of these host factors ( TNPO3 , RanBP2 and Nup153 ) . All three proteins can bind to CA [19] , [27]–[30]; however , their binding to CA does not perfectly correlate with HIV-1 dependence on these host factors [31] . This suggests the presence of an upstream molecule ( s ) that can interact with CA to determine the nuclear entry pathway taken by the PIC . One such molecule is cyclophilin A ( CypA ) because it directly binds to incoming viral capsids [32] , [33] and modulates HIV-1 utilization of TNPO3 , RanBP2 and Nup153 [19] , [26] , [34] . CPSF6 , which also carries a CA-binding domain , is another molecule that may be important in regulating HIV-1 nuclear entry . In fact , CPSF6 binding to CA more strongly correlates with TNPO3 utilization of HIV-1 than CypA binding [18] , [35] , [36] . Thus , one attractive hypothesis proposed by Price et al is that CPSF6 enables HIV-1 to utilize host factors [36] , such as TNPO3 , RanBP2 and Nup153 , thus setting the course of HIV-1 nuclear entry . However , the precise role of CPSF6 in HIV-1 replication and its physiological relevance have yet to be firmly established , partly because: 1 ) There is no experimental evidence for the binding of CA to endogenous CPSF6 . 2 ) CPSF6 knockdown has little effect on wild-type ( WT ) HIV-1 infection in vitro [18] . 3 ) Finally , TNPO3 depletion may keep CPSF6 in the cytoplasm , consequently creating an artificial situation where HIV-1 is effectively blocked [35] , [37] . Taken together , it is controversial whether CPSF6 has any direct role in HIV-1 infection . As mentioned , HIV-1 has an exceptional ability to infect non-dividing cells as efficiently as dividing cells . However , previous studies discovered unusual HIV-1 CA mutants that lost this phenotype and became dependent on cell cycle progression for their infection [38] , [39] . One key observation is that most of such cell cycle-dependent CA mutants are blocked by CypA [39]–[41] . The mechanism by which these mutants become sensitive to an inhibitory function of CypA has been controversial , but one report hinted a co-factor that acts together with CypA in thwarting infection by these CA mutants [42] . Here we provide evidence that endogenous CPSF6 acts together with CypA to exert a detrimental effect on such cell cycle-dependent HIV-1 variants , which include naturally occurring HIV-1 CTL escape mutants in individuals carrying HLA-B27 [43] . HIV-1 can escape from CPSF6-mediated restriction by changing or eliminating CPSF6 binding during in vitro viral evolution . Surprisingly , the escape pathway in vivo is entirely different from in vitro , because in vivo evolution recurrently selects for viruses that strictly maintain CPSF6 binding , suggesting an opposing , beneficial role of CPSF6 for HIV-1 replication in vivo . CPSF6 diminishes the ability of restricted viruses to utilize TNPO3 , RanBP2 and Nup153 . However , in vivo evolution restores this ability . Thus , the preferential selection of CPSF6-dependent viruses in vivo may be beneficial by fostering the utilization of HIV-1 cofactors involved in nuclear entry .
In the present study , we utilized CA mutations localized in the defined CPSF6-binding pocket to explore the function of endogenous CPSF6 for HIV-1 infection . We noticed that A105T , one such CA mutation within the CPSF6-binding pocket [36] , was identified as a compensatory mutation to rescue the T54A mutant [44] . The T54A mutant is defective even in proliferating cells; however , as one of the cell cycle-dependent CA mutants , its infection is more severely impaired in non-dividing cells [44] . This observation suggested to us that CPSF6 binding may affect infection of non-dividing cells and that if A105T altered this binding it would rescue cell cycle-dependent CA mutants . To determine whether A105T indeed prevents CA from binding to CPSF6 , we first used a restriction assay based on a fragment of CPSF6 , called CPSF6-358 . In this assay , restriction of infection occurs when viral capsids are recognized by CPSF6-358 [18] . While the T54A CA mutant was as sensitive to CPSF6-358 restriction as WT CA , we find that the introduction of A105T in the T54A CA ( Fig . 1A ) , as well as in WT CA ( data not shown ) , rescues infectivity ( p<0 . 01 ) , suggesting that this A105T mutation prevents binding of CPSF6-358 . Consistent with this , virions carrying the A105T mutation were unable to abrogate CPSF6-358 restriction when increasing amounts of abrogating virions were used to infect CPSF6-358 expressing cells that were co-infected with the tester WT virus ( Fig . S1 ) . Finally , to directly test whether A105T inhibits CPSF6 binding , we carried out an in vitro binding assay using recombinant HIV-1 CA tubular assemblies . Incubation of the T54A CA tubes with HeLa cell lysates followed by low-speed centrifugation resulted in cosedimentation of endogenous CPSF6 ( Fig . 1B ) at a level comparable to the WT CA tubes ( Fig . S2 ) . By contrast , T54A+A105T CA tubes brought down one-third of the quantity of CPSF6 as the T54A CA tubes ( Fig . 1C ) , indicating that the A105T substitution reduces the association of CPSF6 with the HIV-1 capsid . These findings provide both genetic and biochemical evidence that the A105T substitution reduces the binding of CA to CPSF6 . To confirm that blocking CPSF6 binding to the HIV-1 capsid rescues infection by T54A , we introduced N74D , a mutation known to prevent CPSF6 binding [18] , into the T54A mutant virus . Similar to A105T , N74D allowed T54A to escape from inhibition by CPSF6-358 ( Fig . 1A ) and reduced binding of CA tubes to CPSF6 in vitro ( Fig . S2 ) Moreover , N74D acted similarly to A105T as reported by Qi et al . [39] with respect to its ability to restore T54A infectivity in aphidicolin-arrested HeLa cells ( Fig . 1D ) and infection of actively proliferating HeLa cells ( Fig . 1D ) . These observations suggested that mutations in the CPSF6-binding pocket restore T54A infectivity by preventing the binding of CA to CPSF6 . As T54A is also inhibited for infection of growth-arrested HeLa cells , we hypothesized that endogenous CPSF6 contributes to inhibition of cell cycle-dependent HIV-1 CA mutants . To test this , we asked whether depleting endogenous CPSF6 would restore infectivity of the T54A mutant virus . CPSF6 depletion by siRNA rendered cells more permissive to infection by T54A ( Fig . 2A , Fig . 2B; p<0 . 002 ) . We confirmed this observation by including non-targeting siRNA as controls ( Fig . 2B ) as well as with another CPSF6-targeting siRNA ( data not shown ) . In contrast , the two double mutants ( T54A+N74D and T54A+A105T ) were not enhanced further by knockdown of CPSF6 ( Fig . 2B ) . Moreover , the wild type virus exhibited only modest increase in infectivity upon CPSF6 depletion . Thus , CPSF6 appears to specifically inhibit infection by the HIV-1 T54A mutant . To determine whether the sensitivity to CPSF6-mediated inhibition is specific to CA mutants that lost the ability to infect non-dividing cells , we examined cell cycle-dependent and –independent CA mutants for their response to CPSF6 knockdown . We depleted endogenous CPSF6 and infected the cells either with CA mutants that become dependent on cell division ( E45A , Q63A/Q67A , A92E , G94D , and Q219A ) or those that are impaired for infection but maintain cell cycle independence for infection ( P38A , E71A , E128A/R132A and R143A ) [38] . Infectivity of all of the tested cell cycle-dependent CA mutants was increased upon CPSF6 depletion to a statistically significant extent ( Fig . 2C and Fig . S3 ) . In contrast , infection by none of the cell cycle-independent CA mutants was affected by CPSF6 depletion ( Fig . 2D ) . Thus , CPSF6 appears to selectively inhibit cell cycle-dependent CA mutants . To address the physiological relevance of the novel inhibitory function of CPSF6 for HIV-1 CA mutants , we extended our study to CTL escape mutations ( R132K/L136M; hereafter called RKLM ) in HLA-B27+ individuals . The R132K mutation occurring at the HLA-B27 anchor position −2 prevents the binding of B27 to the peptide and thus allows escape from HLA-B27-restricted CTL response [45] . We became interested in these CA mutations because Qi et al . [39] showed that infection by HIV-1 encoding the R132K substitution , which shares similar properties with RKLM [43] , is also inhibited by aphidicolin treatment of cells ( Fig . 3A ) . Because of its phenotypic similarity to T54A , we hypothesized that the RKLM virus is also restricted by CPSF6 . In support of this , depletion of endogenous CPSF6 markedly enhanced the infectivity of the RKLM mutant to nearly WT levels ( Fig . 3B; p = 0 . 0002 ) . Moreover , addition of the N74D or A105T substitutions in the CPSF6-binding pocket increased the infectivity of RKLM in non-dividing cells ( Fig . 3A ) . Upon infection with RT-normalized viruses , overall infectivity defects were restored even in dividing HeLa cells ( Fig . 3A; p<0 . 001 ) . The triple mutants RKLM+N74D and RKLM+A105T were no longer sensitive to CPSF6-mediated restriction , as they did not exhibit any increase in infectivity upon CPSF6 depletion ( Fig . 3B ) . We next determined whether endogenous CPSF6 inhibits RKLM in primary CD4+ T cells as a more physiological relevant cell type . Infection by RKLM was reduced 3-fold compared to WT ( Fig . 3C; p = 0 . 03 ) . Addition of mutations in the CPSF6 binding pocket ( N74D or A105T ) almost completely rescued viral infectivity of RKLM in activated primary CD4+ T cells ( Fig . 3C; p<0 . 01 ) . Thus , the antiviral effect of CPSF6 is active against virus carrying naturally occurring mutations in relevant targets of HIV-1 infection in vivo . Previous studies showed that CypA contributes to cell cycle-dependent infection by HIV-1 CA mutants [39]–[41] , [46] , including RKLM [39] . To determine whether CPSF6 and CypA block these viruses independently or together , we investigated their restrictive effects in a condition where one of the two molecules is prevented from interacting with viral capsids . Blocking CypA-CA interactions through a genetic mutation ( P90A ) ( Fig . 3D ) or by addition of CsA ( data not shown ) rendered RKLM less sensitive to the inhibitory function of CPSF6 . As previously reported [43] , infection by RKLM was enhanced by neutralizing the inhibitory action of CypA by blocking CypA-CA interactions ( by either addition of CsA or CypA depletion ) ( Fig . 3E , 3F ) . Moreover , blocking CPSF6-CA interactions through genetic mutations ( N74D and A105T ) rendered RKLM unresponsive to the inhibitory action of CypA ( Fig . 3E , 3F ) . Double knockdown of CypA and CPSF6 did not have an additive effect on the restoration of RKLM ( Fig . 3G ) . Similar observations were obtained with the T54A CA mutant ( Fig . S4 ) , except that T54A was less responsive to CPSF6 depletion . These observations uncover the functional crosstalk between CypA and CPSF6 in thwarting infection of cell cycle-dependent CA mutants . As an initial approach to determine whether CypA promotes CPSF6 binding to the HIV-1 capsid , we performed an in vitro binding experiment in which CA-CypA interactions were prevented by addition of CsA . However , the presence or absence of CsA did not detectably alter the ability of CPSF6 to bind to CA tubes , suggesting that CypA and CPSF6 bind independently to the HIV-1 capsid ( Fig . S4E ) . To gain further insight into the mechanism of the CPSF6-mediated restriction of HIV-1 CA mutants , we adapted the RKLM mutant in immortalized CD4+ T cell lines . Serial passages of the RKLM selected two independent viruses that are highly improved in their replicative capacity . Sequence analysis of the CA region of these two adapted viruses did not reveal any reversion of RKLM to the WT sequence but instead one novel amino acid replacement in each of the two adapted viruses: S41A and T107I . While S41A is a common compensatory mutation for RKLM arising in HLA-B27+ individuals [43] , T107I is of particular interest because it is located within the CPSF6 binding pocket [36] . Introduction of each mutation into the molecular infectious clone of RKLM restores viral replicative capacity of RKLM in CEM cells ( Fig . 4A ) as well as the ability to infect non-dividing cells ( Fig . 4B ) . These two adaptive mutations ( S41A or T107I ) confer resistance to the CPSF6-mediated restriction , as the RKLM virus carrying either of these mutations was not rescued upon depletion of CPSF6 and CypA independently or simultaneously ( Fig . 4C ) . T107 directly participates in binding with V314 of CPSF6 [36] . While neither mutation alone ( S41A or T107I alone ) alters HIV-1 sensitivity to CPSF6-358 restriction ( Fig . S5A ) , we observed that RKLM+T107I is less sensitive to CPSF6-358 restriction than WT virus ( Fig . 5A and S5B; p<0 . 0001 ) . Importantly , the phenotype of RKLM+T107I resembles that of T107A; T107A was shown to have reduced affinity to the short peptide from CPSF6 and remained slightly sensitive to CPSF6-358 restriction [36] . These observations suggest that T107I reduces the binding of CA to CPSF6 . In support of this hypothesis , both HIV-1 CA tubes and HIV-1 CA-NC tubes bearing RKLM+T107I substitutions in CA exhibited reduced binding ( ∼50% ) to CPSF6 in vitro ( Fig . 5B , 5C , 5D , 5E ) , while all of the HIV-1 CA-NC tubes had similar levels of binding to rhesus TRIM5α ( Fig . 5D ) . This reduction in CA binding to CPSF6 also parallels the observation with T54A+A105T , which also exhibits reduced binding to CPSF6 ( Fig . 1B and 1C ) . Therefore , these findings support the idea that inhibiting CA binding to CPSF6 is a common mechanism for cell cycle-dependent HIV-1 CA mutants to overcome CPSF6-mediated restriction . Taken together , we demonstrate that endogenous CPSF6 hampers infection of naturally occurring HIV-1 CA variants , possibly through binding to incoming viral capsids together with CypA . The inhibitory function of CPSF6 was puzzling given the high degree of conservation of CPSF6 binding among diverse lentiviruses [36] , [47] . As N74D , one of the CPSF6-independent viruses , is defective in primary macrophages [19] , [48] , CPSF6 may possess a beneficial function that is required specifically for HIV-1 replication in vivo . To further delineate the functional role of CPSF6 in HIV-1 replication , we exploited the unique pattern of CA evolution in HLA-B27+ subjects [43] . As described above , the RKLM mutations allow viral escape from the CTL response , yet the same RKLM mutation would render the virus sensitive to CPSF6-mediated restriction . The consequences of these two opposing virus-host interactions are an effective control of viremia in HLA-B27+ subjects . However , similar to our in vitro adaptation , in vivo evolution generates late CTL variants that can sustain high levels of viral replication in these individuals [43] . We asked whether and how these late variants escape from CPSF6-mediated restriction . We generated chimeric viruses carrying the entire capsid sequence from five HLA-B27+ subjects infected with HIV-1 [49] ( Fig . S6 ) . All of these CA sequences contain the R132K substitution , which confers escape from CTL response but sensitivity to CypA-mediated restriction [39] , [43] , which is identical to CPSF6-mediated restriction . As described above , one notable property accompanying escape from CPSF6-mediated restriction is the restorated ability to infect non-dividing cells . We find that all of the chimeric viruses , except for SH8127 , retained this ability , as opposed to the “early” CTL escape mutant just carrying RKLM ( Fig . 6A ) . These same four viruses also appeared to acquire resistance to CPSF6-mediated restriction , as they were no longer enhanced by depletion of either CPSF6 or CypA , while like RKLM , SH8127 infectivity was enhanced by CPSF6 depletion ( Fig . 6B ) . Finally , CR0339X , the only replication-competent virus we generated among the four adapted viruses , replicated efficiently in CEM cells , while SH8127 was defective ( Fig . 6C ) . These data suggest that all capsid sequences , except for SH1827 , adapted in vivo by escaping CPSF6-mediated restriction . Notably , CPSF6 escape correlates well with viral loads in vivo . All of the reported subjects had high viral loads ( 179 , 000–750 , 000 viral RNA copies per mL ) [49]; this is in contrast to the subject from which SH8127 was isolated , who exhibited minimal detectable plasmid viral RNA . Previous reports demonstrated that mutations at S41 in the CA sequence restore in vitro replicative capacity to CTL escape variants carrying the signature RKLM mutation [43] . Consistent with this in vitro observation , mutations at S41 are highly associated with increased viral loads in HLA-B27+ individuals [43] , [49] . While all four adapted CA sequences contain mutations at S41 , SH8127 did not bear any genetic change at this position ( Fig . S6 ) . In fact , when S41 was restored in three of the four adapted CA sequences , two of them regained the sensitivity to restriction by CPSF6 ( Fig . S7 ) , underscoring the role of this position in sensitivity to CPSF6-mediated restriction . Taken together , albeit with the small sample size , the sensitivity to CPSF6-mediated restriction and viral loads in HLA-B27+ subjects is tightly correlated . We next asked how these late variants escape from CPSF6-mediated restriction . Our in vitro adaptation experiments showed that RKLM can escape from this restriction by modulating the ability of CA to interact with CPSF6 ( i . e . RKLM+T107I ) . However , none of the four adapted CA sequences selected for in vivo have any mutations at amino acids that participate directly in binding to the CPSF6 peptide [49] ( Fig . S6 ) . In fact , we find that chimeric viruses carrying these capsid sequences were effectively blocked by CPSF6-358 ( Fig . 6D; p<0 . 01 ) , suggesting that they still maintain CPSF6 binding . As described earlier ( Fig . 3E , 3H ) , preventing CA binding to CypA rescues viruses sensitive to CPSF6-mediated restriction . Thus , such sensitive viruses could potentially escape from the CPSF6-mediated restriction by modulating the binding to CypA . However , these four adapted viruses , some of which contain mutations in CypA-binding loop [32] , [50] ( Fig . S6 ) , also retain the binding to CypA , as evident by their sensitivity to the fusion protein between TRIM19 and CypA ( TRIM19-CypA ) ( Fig . 6E; p<0 . 01 ) . While this genetic assay with TRIM19-CypA is convenient to detect the binding of CA to CypA , it may not fully recapitulate physiologically relevant CA-CypA interactions that promote HIV-1 infection . To circumvent this potential issue , functional interaction with CypA was confirmed with the CypA-null Jurkat cell line in which CypA-dependent viruses were all blocked relative to the control Jurkat cell line ( Fig . 6F; p<0 . 05 ) . As described above , all four adapted viruses contain compensatory mutations at S41 that are almost invariably found among other late variants in HLA-B27+ subjects . The RKLM mutant carrying S41A , which was incidentally isolated in our in vitro adaptation , also retained binding to CPSF6 ( Fig . 5A , Fig . S5B , Fig . 5D , Fig . 5E ) . We noticed a trend in which RKLM+S41A binds to CPSF6 ∼7% weaker than RKLM; however , this was not supported by statistical significance ( Fig . 5E ) . Interestingly , the same mutant ( RKLM+S41A ) was less dependent on CypA that WT for its infection ( Fig . 6F ) . Altogether , these results reveal two opposing CA-dependent functions of CPSF6 during HIV-1 replication in vivo . CPSF6 exerts deleterious effects on the early CTL escape variants in HLA-B27+ subjects . Such a restricted virus has the potential to escape from CPSF6 by modulating CPSF6 binding ( Fig . 5 ) . However , in vivo evolution selected for viruses that escape from CPSF6-mediated restriction yet retain CPSF6 binding . This observation supports the idea that CPSF6 is beneficial for HIV-1 replication in vivo . CPSF6 binding is strictly maintained in vivo but is dispensable for HIV-1 replication in vitro . Given that activated CD4+ T cells are the major cell type to produce HIV-1 in vivo , we investigated the requirement of CPSF6 for HIV-1 replication in primary CD4+ T cells by studying how the RKLM mutant will evolve in primary CD4+ T cells . Three independent experiments for experimental adaptation of the RKLM virus in this cell type generated three genetic changes ( V86A , A92V and Q112H ) in the viral capsid sequences without reversion of the parental RKLM mutations . We examined the sequence of only the CA region of these adapted viruses , but each of these three mutations , when separately introduced into the parental RKLM mutant , restored infectivity to RKLM in both CEM cells and primary CD4+ T cells ( Fig . 7A ) . These three mutations confer resistance against CPSF6-mediated restriction , as the viruses containing each of the mutations regained the ability to infect non-dividing cells ( Fig . 7B ) . Furthermore , they were not significantly rescued by CPSF6 or CypA knockdown , relative to RKLM ( Fig . 7C ) . Two of these mutations , V86A and A92V , are located within the CypA binding loop [32] . In fact , RKLM carrying each of these three mutants was less sensitive to restriction by TRIM19-CypA than WT ( Fig . 7D , p<0 . 05 ) . However , their infectivity was impaired in CypA-null Jurkat cells similar to that by WT ( Fig . 7E ) . More importantly , we find that they retain the ability to interact with CPSF6 by the CPSF6-358 restriction assay ( Fig . 7F ) . Thus , ex vivo escape from CPSF6-mediated restriction takes a similar course as in vivo escape . In both settings , HIV-1 strictly maintains CPSF6 binding as opposed to in vitro adaptation . What drives recurrent selection of CPSF6 binding in vivo ? Price et al . proposed a new role for CPSF6 in the utilization of host factors implicated in HIV-1 nuclear entry , such as TNPO3 , RanBP2 and Nup153 [36] . To explore this idea , we examined pre-adapted and adapted viruses for their dependence on these host molecules by depleting these gene products by RNAi . We find that the “early variants” ( RKLM or the pre-adapted CTL escape mutant SH8127 ) were less sensitive or completely insensitive to depletion of these molecules ( Fig . 8; p<0 . 05 when compared with WT ) , indicating that CPSF6-mediated restriction eliminates the ability to utilize these molecules . In contrast , all the adapted viruses in vivo ( “late variants” ) were blocked at levels comparable to the WT virus ( Fig . 8; p<0 . 05 ) . Therefore , escape from CPSF6-mediated restriction in vivo maintains CPSF6 binding and restores the ability to utilize these host molecules ( Fig . 8 ) . Again , all three ex vivo-adapted viruses mirror the in vivo-adapted viruses , since they were all dependent upon these nuclear entry factors ( Fig . 8; p<0 . 05 ) . Interestingly , RKLM+T107I , which is less sensitive to CPSF6-358 and resistant to CPSF6-mediated restriction , also regains the ability to utilize these nuclear entry factors ( Fig . 8; p<0 . 05 ) . In summary , these findings indicate that in vivo escape from CPSF6-mediated restriction that retains CPSF6 binding is strongly correlated with recovery of utilization of TNPO3 , RanBP2 and Nup153 .
There is mounting evidence to suggest that CPSF6 plays a critical role in HIV-1 replication [18] , [35] , [36]; however , its exact function remains to be defined . In this study , we showed that depletion of endogenous CPSF6 rescues infection by HIV-1 CA mutants that are impaired for infection of non-dividing cells ( Fig . 2 , 3 ) . We also found that reducing CPSF6 binding by CA mutations in the CPSF6-binding pocket restores infectivity to HIV-1 CA mutants that are sensitive to CPSF6-mediated restriction ( Fig . 1 , 5 ) , suggesting that CPSF6-mediated restriction depends on direct binding of endogenous CPSF6 to incoming viral capsids . Previous studies , which employed in vitro binding assays , restriction by CPSF6-358 , and cytoplasmic localization of CPSF6 by TNPO3 knockdown [18] , [35] , [36] , indicate that CPSF6 can directly bind to the viral capsid . Our current data complement these observations and support the idea of the direct interaction between endogenous CPSF6 and incoming capsids in a physiological setting . We observed that infection by CA mutants restricted by CPSF6 is less dependent or independent of the nucleoporins RanBP2 and Nup153 . As these host proteins are components of the nuclear pore complex , we suggest that CPSF6 engages the viral capsid in the cytoplasm . In this respect , CPSF6 appears to resemble cyclophilin A ( CypA ) , as CypA also binds to incoming viral capsids in the cytoplasm [32] , [33] . Our data also reveal the functional crosstalk between CPSF6 and CypA . Specifically , CA mutants sensitive to CPSF6-mediated restriction are also inhibited by CypA [39] , [41]–[44] , [46] , [51] , [52] . Expression of CPSF6 is essential for CypA-mediated restriction , which parallels CPSF6-mediated restriction ( Fig . 3 , Fig . S4 ) . Mutations in CA that result in cell cycle-dependent infection may alter the normal course of viral uncoating . In this model , CypA and CPSF6 may exacerbate such an uncoating defect exhibited by cell cycle-dependent CA mutants . This is consistent with a recent finding that reveals core stabilization as a mechanism of restriction of the WT virus by cytoplasmic CPSF6 [35] , [37] . CypA has also been reported to control HIV-1 uncoating both in vitro and in target cells [34] , [46] . Other studies suggest a potential direct role for TNPO3 in HIV-1 uncoating [27] , [34] , thus it may be possible that TNPO3 is also involved in the inhibitory action of CPSF6 and CypA for cell cycle-dependent HIV-1 CA mutants , although this hypothesis appears inconsistent with our observation that TNPO3 depletion did not enhance infectivity of RKLM ( Fig . 8 ) . The antiviral action of CPSF6 is physiologically relevant to HIV-1 replication in vivo , because we find that CPSF6 inhibits infection of early CTL escape variants in HLA-B27+ individuals ( Fig . 3 , 6 ) . Moreover , our data reveal a positive correlation between sensitivity to CPSF6-mediated restriction and effective suppression of viral replication in vivo . Namely , one capsid ( SH8127 ) susceptible to CPSF6 restriction originated from a subject with very low viral loads ( 50 viral RNA copies per mL ) [49] , whereas those resistant to CPSF6 restriction were derived from individuals harboring much higher viral loads ( 179 , 000–750 , 000 viral RNA copies per mL ) [49] . Furthermore , compensatory mutations at S41 that confer resistance to CPSF6-mediated restriction are significantly associated with higher viral loads in HLA-B27+ individuals [43] , suggesting that escape from CPSF6-mediated restriction is essential for robust HIV-1 replication in these individuals . Naturally , this parallels the observation made by Schneidewind et al . who found that CypA is inhibitory towards early CTL escape variants [43] . Interestingly , mutations associated with HLA-B27 also increase sensitivity of HIV-1 to human TRIM5α [53] . Hence , inhibitory functions of these CA-binding host proteins contribute to the constraints that render CA immunologically vulnerable [54] . The inhibitory function of CPSF6 was somewhat unexpected given the conservation of the CPSF6-binding pocket in CA proteins from diverse primate lentiviruses [36] , [47] . However , the current study also provides evidence for an opposing , beneficial role of CPSF6 for HIV-1 replication in vivo . This was possible because of the unique pattern of HIV-1 CA evolution in HLA-B27+ subjects . Our data reveal the strong selective pressure for maintaining CPSF6 binding to CA in these subjects ( Fig . 6 ) . This is significant because this strict preservation of CPSF6 binding in vivo occurs despite the opposing pressure to reduce CPSF6 binding in order to escape from its inhibitory activity , as shown in vitro ( Fig . 5 ) . As described above , HIV-1 evolution in HLA-B27+ individuals takes one major escape pathway utilizing substitutions at S41 to evade CPSF6-mediated restriction . Thus , despite the small sample number in this study , it is likely that the preservation of CPSF6 binding is an inevitable requirement for adaptation of HIV-1 in these individuals . Therefore , these findings suggest that CPSF6 has an alternative role that is advantageous to HIV-1 replication in vivo . What is the driving force to select for CA that binds CPSF6 in vivo ? One proposed function of CPSF6 during HIV-1 infection is that CPSF6 controls the nuclear entry pathway for HIV-1 [36] . Consistent with this model , the strict preservation of CPSF6 binding in vivo was highly correlated with utilization of TNPO3 , RanBP2 and Nup153 ( Fig . 8 ) . Namely , we observed that early CTL escape variants that lack adapted mutations at S41 are less dependent on TNPO3 and completely independent from RanBP2 and Nup153 , but all of the late adapted viruses regained the ability to utilize these three cellular molecules . These observations suggest that utilizing TNPO3 , RanBP2 and Nup153 is a major selective pressure for the maintenance of CA binding to CPSF6 in vivo . We propose that CPSF6 interactions with cell cycle-dependent mutants prevent the viral core from accessing the cellular nuclear entry machinery , and that mutations reducing CPSF6 binding relieve the impairment . As CPSF6 also binds the wild type HIV-1 capsid but does not inhibit infection by the wild type virus , it is possible that the RKLM substitutions weaken the interaction of the viral capsid with the nuclear pore , rendering the capsid sensitive to masking by CPSF6 . Interestingly , while T107I and N74D substitutions in CA appear to rescue RKLM by reducing the binding to CPSF6 , the S41A substitution rescued RKLM from CPSF6 inhibition without significantly altering the extent of CPSF6 binding ( Fig . 5 ) . Thus , S41A may relieve an uncoating defect induced by CPSF6 binding to the RKLM capsid , or it may enhance subsequent interactions with the nuclear pore proteins thereby circumventing the CPSF6 block . Studies of the interaction of the RKLM capsid with nucleoporins should help address this issue . CPSF6 binding by HIV-1 is strictly conserved in vivo but flexible in vitro . For instance , in vitro escape of cell cycle-dependent CA mutants from the antiviral function of CPSF6 results in mutations ( A105T and T107I ) in the CPSF6 binding pocket [44] ( Fig . 4 ) , which reduce the CA-CPSF6 interactions ( Fig . 1 , 5 ) [35] . Similarly , HIV-1 evasion from an antiviral form of CPSF6 selected for the first CPSF6-binding deficient mutation ( N74D ) [18] . The difference between in vitro and in vivo in their ability to accommodate genetic variability within the CPSF6-binding pocket is somewhat similar to the recent observation that some CA mutations with high in vitro fitness did not occur in natural HIV-1 subtype B populations [55] . However , the virus carrying the N74D mutation is severely attenuated in macrophages [19] , [48] , pointing to the possibility that the dependence of HIV-1 on CPSF6 may be cell type-specific . These observations may be clinically relevant . PF-3450074 is a recently described anti-HIV-1 compound that binds to the CPSF6-binding domain of the capsid . HIV-1 was shown to gain resistance against this compound in vitro by allowing mutations within the same binding domain [56] , [57] , but the escape barrier for such drugs would likely be high in HIV-1-infected individuals , since the CPSF6-binding domain of CA is immutable in vivo . In this study , we demonstrated that the host protein CPSF6 possesses two opposing functions for HIV-1 replication in vivo . Viral evolution in HLA-B27+ subjects suggests a beneficial function of CPSF6 for HIV-1 replication in vivo that outweighs its inherently deleterious potential . This unique CPSF6-HIV interplay illustrates the complex nature of host-pathogen interactions . Physical contact of viral molecules with cellular components is one fundamental feature for viral hijacking of the host machinery , while at the same time host factors may become detrimental to viral replication . In this respect , the CPSF6-CA interaction may not be completely unique as a recent comprehensive study revealed additional HIV-human protein interactions that can attenuate HIV-1 replication [58] .
Molecular infectious clones based on the LAI strain were generated by introducing mutations into either a replication-competent pBru3oriGFP3 backbone [11] , one that is Env-defective , or both using standard cloning procedures . Entire sequences of the capsid-encoding segment of HIV-1 Gag from five different infected individuals carrying the HLA-B27 haplotype , which were previously reported by Wang et al . [49] , were synthesized ( Life Technologies ) and cloned into the above-mentioned molecular infectious clones . The E2-Crimson gene was cloned into the nef position of the wild-type env-deficient HIV-1 clone to generate pBru3ori-ΔEnv-Crimson . An HIV-1 Gag-Pol expression vector , pCRV1-Gag-Pol ( LAI ) , was a derivative of pCRV1-Gag-Pol ( gift from the Hatziioannou and Bieniasz labs ) constructed by replacing the most of the Gag encoding sequence with that from LAI ( from the Gag start codon to the unique ApaI site in LAI ) . Various CA mutations were introduced into this derivative by using standard cloning techniques . The gene depletion vector pLKO . 1-Crimson-RanBP2 that contains shRNA targeting RanBP2 [19] based on the previous target sequence was generated by replacing the puromycin resistance gene of the parental pLKO . 1 vector with the E2-Crimson gene amplified from the pTEC19 plasmid ( Addgene ) . An HIV-1 CA-NC expression vector , pWISP96-18 ( a kind gift of Wesley Sundquist ) [59] , was used as a template to introduce different HIV-1 CA mutations by overlapping PCR . HeLa cells were cultured in DMEM ( Cellgro ) and supplemented with 10% FBS ( Cellgro ) and 1× penicillin-streptomycin ( P/S , Cellgro ) . Immortalized suspension cells ( MT4 , CEM , Jurkat ) were cultured in RPMI ( Cellgro ) and supplemented with 10% FBS ( Sigma ) , 1× P/S and 1× 2-glutamine ( Cellgro ) . PBMCs were isolated from whole blood obtained from anonymous blood donors ( New York Blood Center ) using standard Ficoll ( Cellgro ) procedures . Primary CD4+ T cells were isolated using the Human CD4+ T Cell Enrichment Kit per manufacturer instructions ( Easysep ) . Primary CD4+ T cells were activated using Dynabeads Human T-Activator CD3/CD28 ( Gibco ) and cultured in suspension cell medium supplemented with 30 units per mL of IL-2 ( Peprotech ) . JurkatΔCypA cells [60] and HeLa cells overexpressing either TRIM19-CypA cells [40] or CPSF6-358 ( gift by V . KewalRamani ) [18] were previously described . HeLa cells were plated at 5×105 cells per well of a 6-well plate and transfected with 30 pmol siRNA using Lipofectamine RNAiMAX ( Invitrogen , 13778 ) or with the transfection reagent alone ( control ) . Non-transfected cells were used as controls in all experiments . Non-targeting siRNA were also used in certain experiments as additional controls . We found no significant difference in viral infectivity between these controls . Specific siRNA used for knockdowns were: anti-CPSF6 ( Thermo Scientific J-012334-11 , J-012334-09 ) , anti-CypA ( GAUGAACUUCAUCCAGACUUU ) , anti-TNPO3 ( Thermo Scientific L-019949-01-0010 ) and anti-Nup153 ( GGACUUGUUAGAUCUAGUUUU ) . The following day the transfection was repeated with the same procedure . Four hours post-transfection , cells were seeded at 5×105 cells per 96-well plate for infection . For RanBP2 knockdown experiments , HeLa cells plated at 5×105 cells per well of a 6-well plate were infected with VSV-G-pseudotyped Crimson reporter viruses carrying RanBP2-targeting shRNA . Effects of gene depletion were analyzed by western blotting ( Fig . S8 ) . Two days after infection , 5×105 cells were plated onto 96-well plates for a second infection with GFP reporter viruses . All viruses were generated using 293T cells as previously described [11] using polyethylenimine ( PEI , PolySciences ) as the transfection reagent . pBru3oriGFP3ΔEnv viruses were pseudotyped with the VSV-G envelope . Viruses were normalized using the Lenti RT Activity Kit ( Cavidi ) . All infections using HeLa cells were performed at 5×105 cells per plate . HeLa cells treated with 2 µg per ml of aphidicolin ( Aph , Sigma ) were plated at 1 . 5×106 cells per plate . Cyclosporine A ( CsA , Sigma ) was added to the culture during infection at 2 µM . Immortalized T Cell lines and primary CD4+ T cells were plated at 2 . 5×105 cells per mL and 1×106 cells per mL , respectively . Virus infectivity was examined by measuring crimson- and/or GFP-positive cells using a BD LSRII Flow Cytometer . Abrogation experiments were done by co-infecting CPSF6-358-expressing HeLa cells with a fixed amount of HIV-1 WT virus ( encoding the E2 crimson ) and increasing amounts of abrogating viruses carrying various CA mutations . We used the amount of Crimson reporter virus such that infection with a restricted WT virus generates 0 . 2–1% Crimson-positive cells without restriction-abrogating viruses . Abrogating particles were generated by transfecting three plasmid DNA ( packageable GFP-encoding HIV-1 vector , HIV-1 Gag-Pol expression vector and VSV-G-encoding vector ) [61] . Two days after infection , crimson-positive cells were measured as described above . MT4 cells ( 1 . 25×105 per well of a 24-well plate ) were infected with replication-competent virus carrying the RKLM mutation ( 3 ng of reverse transcriptase ) . Supernatant harvested at 16 days post-infection ( dpi ) was used to infect fresh MT4 cells . DNA was extracted from acutely infected MT4 cells at 1 dpi using the DNeasy Blood & Tissue Kit ( Qiagen ) . PCR was performed to amplify the segment of the Gag gene encoding the CA protein using primers ( HIV-1-LAI-CA-Fwd 5′-GCACAGCAAGCAGCAGCTGACACAGG; HIV-1-LAI-CA-Rev 5′-GCCTCTTTGCATCATTATGGTAGC ) . The amplified fragments were subjected to direct sequencing with the same PCR primers ( Genewiz ) . The MT4-adapted CA sequence had a nucleotide switch at CA C310T ( based on the LAI strain: K02013 ) , which leads to a threonine-to-isoleucine change at residue 107 of the viral CA protein ( T107I ) . CEM cells were infected at 1 . 25×105 cells per well of a 24-well plate with the RKLM virus ( 13 ng RT ) in the presence of 2 µM CsA , which was added to augment the initial infection . The adapted virus was harvested at 21 dpi and analyzed by sequencing of the CA-encoding segment . The CEM-adapted sequence had a T-to-G nucleotide change at 121 in the viral capsid sequence , resulting in to a serine-to-alanine change at 41 of the viral CA protein . All primary CD4+ T cell adaptation experiments were carried out using isolated CD4+ T cells activated with CD3/CD28 beads . Each independent experiment was performed using the same donor , with a total of 3 separate donors . CD4+ T cells were infected at 5×105 cells per well of a 24-well plate with the RKLM virus ( 30 ng RT ) . At 3 dpi , infected CD4+ T cells were mixed one-to-one with freshly isolated and activated CD4+ T cells from the same donor . This process was repeated until the proportion of virus-infected cells , as judged by GFP positivity , reached 15% . Cell-free supernatants were then harvested , passaged further , collected once GFP positivity reached 15% and used to infect fresh MT4 cells . 1 dpi , DNA was extracted and analyzed by sequencing of the CA-encoding segment . In each of the three experiments , we found the following substitutions: V86A was a nucleotide switch at CA T257C , A92V was a nucleotide switch at CA C275T and Q112H was a nucleotide switch at CA G336T . Protein levels were determined using standard laboratory western blot protocols . The primary antibodies used were: anti-CPSF6 ( Proteintech 15489-1-AP ) , anti-CypA ( Thermo Scientific PA1-025 ) , anti-TNPO3 ( Abcam ab54353 ) , anti-RanBP2 ( Abcam ab2938 ) , anti-tubulin ( Sigma-Aldrich T6074 ) . Protein levels of Nup153 were determined using anti-nuclear pore complex proteins antibody ( a gift from the Blobel laboratory ) . The secondary HRP-conjugated antibodies used were anti-rabbit ( Invitrogen 65-6120 ) and anti-mouse ( Santa Cruz sc-2005 ) . Western blotting of capsid-binding assays was performed by using fluorescent secondary antibodies and signals were quantitated with a LI-COR Odyssey scanner . Differences in infectivity between different conditions ( e . g . between control and knockdown , between WT and mutants ) were examined by a paired Student t-test . P-values of 0 . 05 or less were considered statistically significant . | The viral capsid ( CA ) protein of HIV-1 determines both the ability to infect non-dividing cells and the utilization of host factors implicated in nuclear entry . Understanding how CA controls these two properties is critical . CPSF6 , a CA-interacting host protein , may be important for these properties but its precise role remains unclear . Here we provide direct evidence for the involvement of endogenous CPSF6 during HIV-1 infection . We found that CPSF6 blocks CA mutants that are impaired for infection of non-dividing cells . This CPSF6-mediated inhibition also targets early escape variants that arise in HIV-1 infected HLA-B27+ patients . Moreover , this CPSF6-mediated inhibition , together with robust CTL response , appears to be critical for viral suppression , because viruses derived after late viral breakthrough in these individuals were no longer sensitive to the antiviral activity of CPSF6 . However , we also report indirect evidence for a potentially beneficial role for CPSF6 in HIV-1 replication , because escape from this inhibition in vivo was paradoxically accompanied by a strict preservation of the CPSF6 binding pocket . These results highlight the unique characteristics of the HIV-CPSF6 interactions in which CPSF6 can be either beneficial or detrimental for viral replication in a CA-specific manner . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunodeficiency",
"viruses",
"viral",
"preintegration",
"complex",
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"viral",
"immune",
"evasion",
"virology",
"biology",
"microbiology",
"viral",
"replication",
"viral",
"evolution"
] | 2014 | In Vivo Functions of CPSF6 for HIV-1 as Revealed by HIV-1 Capsid Evolution in HLA-B27-Positive Subjects |
HIV-1 subtype B replication in the CNS can occur in CD4+ T cells or macrophages/microglia in adults . However , little is known about CNS infection in children or the ability of subtype C HIV-1 to evolve macrophage-tropic variants . In this study , we examined HIV-1 variants in ART-naïve children aged three years or younger to determine viral genotypes and phenotypes associated with HIV-1 subtype C pediatric CNS infection . We examined HIV-1 subtype C populations in blood and CSF of 43 Malawian children with neurodevelopmental delay or acute neurological symptoms . Using single genome amplification ( SGA ) and phylogenetic analysis of the full-length env gene , we defined four states: equilibrated virus in blood and CSF ( n = 20 , 47% ) , intermediate compartmentalization ( n = 11 , 25% ) , and two distinct types of compartmentalized CSF virus ( n = 12 , 28% ) . Older age and a higher CSF/blood viral load ratio were associated with compartmentalization , consistent with independent replication in the CNS . Cell tropism was assessed using pseudotyped reporter viruses to enter a cell line on which CD4 and CCR5 receptor expression can be differentially induced . In a subset of compartmentalized cases ( n = 2 , 17% ) , the CNS virus was able to infect cells with low CD4 surface expression , a hallmark of macrophage-tropic viruses , and intermediate compartmentalization early was associated with an intermediate CD4 entry phenotype . Transmission of multiple variants was observed for 5 children; in several cases , one variant was sequestered within the CNS , consistent with early stochastic colonization of the CNS by virus . Thus we hypothesize two pathways to compartmentalization: early stochastic sequestration in the CNS of one of multiple variants transmitted from mother to child , and emergence of compartmentalized variants later in infection , on average at age 13 . 5 months , and becoming fully apparent in the CSF by age 18 months . Overall , compartmentalized viral replication in the CNS occurred in half of children by year three .
Human immunodeficiency virus type 1 ( HIV-1 ) infection of the central nervous system ( CNS ) can occur shortly after transmission , and compartmentalized HIV-1 variants , genetically distinct from the virus in the blood , can be detected in the cerebrospinal fluid ( CSF ) in some individuals throughout the course of infection . Detection of compartmentalized CSF viral populations during primary infection suggests that compartmentalization can occur early in the absence of overt neurological symptoms [1] . Extensive compartmentalization of HIV-1 has been shown to be a strong indicator of HIV-1 neuropathogenesis contributing to HIV-1-associated dementia ( HAD ) [2] , [3] , [4] , [5] , while intermediate levels of compartmentalization are associated with either an asymptomatic state or less severe forms of HIV-1-associated neurological disease [1] , [3] . Although a longitudinal link has not been made , these results raise the possibility that early detection of compartmentalized CSF variants may identify subjects with a higher risk of developing HIV-1-associated neurological complications . Compartmentalized HIV-1 subtype B CNS populations can be either CCR5 ( R5 ) -using T cell-tropic or macrophage-tropic [6] , [7] , [8] , [9] , [10] . Macrophage-tropic HIV-1 variants are characterized by the ability to infect cells with low CD4 surface expression [11] , [12] , [13] , are poorly represented in the blood [9] , are not transmitted [14] , [15] , and decay slowly following initiation of highly active antiretroviral therapy ( HAART ) [16] , [17] , unlike the rapid decay of virus replicating in activated CD4+ T cells within the blood [16] , [18] , [19] . While extensive research has been conducted on HIV-1 subtype B CNS infections , little is known about CNS compartmentalization of HIV-1 subtype C ( the most common subtype worldwide ) or the ability of HIV-1 subtype C to evolve to use low levels of CD4 for entry . Previous studies have reported significant differences in the subtype B and C envelope glycoproteins and suggested that subtype C may be less neuropathogenic than subtype B [20] , [21] , [22] . Additional studies on HIV-1 subtype C CNS infections are needed to improve our knowledge on HIV-1 subtype C pathogenesis , CNS compartmentalization and entry tropism . Information on the genetic and phenotypic characteristics of HIV-1 within the CNS of infants and young children is scarce . HIV-1 CNS disease is often an AIDS-defining illness in children [23] , [24] , [25] , and this implies that early infection of the CNS may be important in the pathogenesis of HIV-1 infection in infants [25] . Understanding the dynamics of pediatric HIV-1 CNS replication is thus of critical importance . We examined HIV-1 variants in 43 ART-naïve Malawian children aged three years or younger to determine the viral genotypes and phenotypes associated with HIV-1 subtype C pediatric CNS infection . We observed intermediate , minor compartmentalization in 25% and distinct , compartmentalized CSF variants in 28% of children . Older age and a higher CSF/blood viral load ratio were associated with CNS compartmentalization , with genetic evidence suggesting outgrowth of the compartmentalized variant starting at around 13 months of age . Transmission of multiple variants had occurred in 5 children , of which 4 had one variant sequestered within the CNS , consistent with early stochastic colonization of the CNS by the virus . Finally , we showed that genetically compartmentalized R5 virus with the ability to infect cells with low CD4 surface expression , a hallmark of macrophage-tropic viruses , can evolve in HIV-1 subtype C CNS infection in young children , although this was not common in the first three years of life . Overall we found that by 3 years of age , 50% of children infected with HIV-1 subtype C had virus independently replicating in the CNS .
We examined viral populations in paired peripheral blood and CSF samples collected from 43 HIV-infected Malawian children presenting with either neurodevelopmental delay or acute neurologic symptoms ( neurodevelopmental delay will be examined in a separate study ) . Subjects were infected with HIV-1 subtype C , were ART naïve , and ranged in age from 3 to 35 months . Blood viral loads ranged from 4 , 344 to >80 , 000 , 000 copies/mL and CSF viral loads ranged from 56 to 4 , 745 , 000 copies/mL . The virological characteristics for each subject are summarized in Table 1 . Infection is assumed to be by a vertical route but the fractions infected prepartum , intrapartum , or postpartum are not known . To asses HIV-1 genetic compartmentalization , cDNA templates were generated from extracted blood and CSF viral RNA and used in single genome amplification ( SGA ) of the full-length viral env gene [26] , [27] , [28] , [29] , [30] . A mean of 19 amplicons were analyzed per sample . The sequence of the entire env gene was determined for each amplicon and phylogenetic analysis was completed . Compartmentalization was assessed visually and statistically using the Slatkin-Maddison test [31] . CNS compartmentalization was defined by a Slatkin-Maddison P value<0 . 05 and a genetically distinct CSF population with a bootstrap value ≥40; we used a relatively low bootstrap value in this exploratory analysis because of the overall low diversity of the viral population in these children . Intermediate populations were defined by a Slatkin-Maddison P value>0 . 05 but with visual evidence of a minor CSF subpopulation of ≥4 CSF amplicons and a bootstrap value ≥40 . Equilibrated populations were defined by a Slatkin-Maddison P value>0 . 05 and no evidence of a minor or major CSF population ( Table 1 ) . Phylogenetic trees were also examined for clonal amplification . Clonally amplified lineages were defined as having short branch lengths in the neighbor-joining phylogenetic tree with bootstrap values ≥99 and a clade of ≥3 variants . These lineages signify the recent amplification of identical or nearly identical variants . Twenty out of 43 subjects ( 46% ) had equilibrated viral populations in their blood and CSF ( Table 1 and Figure 1A ) . For these subjects , sequences from the two compartments were well mixed and the CSF sequences were not genetically distinct from those of the blood . Two of these subjects did , however , display evidence of minor clonal amplification in the CSF . In 11 out of 43 subjects ( 26% ) , an intermediate condition existed where the peripheral blood and CSF HIV-1 populations were not uniformly equilibrated and contained a minor CSF subpopulation ( Table 1 and Figure 1B ) . Six of these intermediate subjects displayed evidence of clonal amplification . We hypothesize that the minor CSF population may indicate a precursor population within the CNS with the potential to expand into a compartmentalized population as infection progresses ( see relationship with age below ) . Significant genetic compartmentalization was detected between the blood and CSF populations in 12 out of 43 subjects ( 28% ) ( Table 1 and Figure 1C ) indicating the presence of an independent , autonomously replicating viral population within the CNS . In these subjects , the virus in the CSF was genetically distinct from the virus in the blood . In 5 compartmentalized subjects , evidence of clonal amplification was observed . Therefore , within the first three years of HIV-1 subtype C pediatric infection , significant genetic compartmentalization can be observed . Relationships between subject characteristics and compartmentalization were assessed using the Mann-Whitney test . Older children were more likely to have compartmentalized CSF variants when compared to equilibrated ( P = 0 . 05 ) and intermediate subjects ( P = 0 . 005 ) ( Figure 2A ) . As the majority of vertical transmission occurs early , either in utero , at delivery or during the first months of breast feeding [32] , older age provides a longer period of time for viral variants to become established within the CSF . No relationship was observed between the blood and CSF viral loads and subject classifications ( Figure S1 ) . However , a higher CSF/blood viral load ratio was significantly related to compartmentalization when compared to equilibrated ( P = 0 . 02 ) or intermediate subjects ( P = 0 . 001 ) ( Figure 2B ) , consistent with the occurrence of local replication and expansion of viral populations within the CNS independent from the peripheral blood . Thus , compartmentalization most often appears subsequent to transmission and is associated with a higher CSF/blood viral load ratio , representing virus produced locally in the CNS . Bayesian Evolutionary Analysis by Sampling Trees ( BEAST ) [33] was used to estimate the time to most recent common ancestor ( TMRCA ) of the entire viral populations and the compartmentalized populations ( Table 1 ) . For the majority of the subjects , transmission was predicted to have occurred at birth ±6 months , as shown by the good concordance between the TMRCA and the age of the child at the time of sampling ( Figure 3A ) . For 5 subjects , the predicted TMRCA was substantially higher than the age of the child , which , based on further analysis ( discussed below ) , was probably due to multiple transmitted viruses . As the majority of subjects were probably infected at or around the time of birth , we were able to depict the occurrence of compartmentalization as a function of time ( Figure 3B ) . Before age 18 months , the populations in half the children were equilibrated , with more intermediate populations than compartmentalized in the remaining half . After age 18 months , about half of the children continued to have equilibrated populations while in the remainder , compartmentalized populations were now much more prevalent than intermediate populations . These data further support a potential transition over time in one-half of the children to increasing CNS compartmentalization in the absence of antiretroviral therapy . For 5 subjects for whom the predicted TMRCA was substantially greater than the age of the subject ( Figure 3A ) , further analysis revealed the presence of multiple transmitted viruses ( Table 1 ) . For one compartmentalized subject ( 3002; age 5 months ) , phylogenetic analysis revealed a deep bifurcation separating two distinct viral populations , one comprised almost exclusively of CSF variants , and the other comprised primarily of blood variants ( Figure 4A ) . Sequence analysis demonstrated that the viral populations were genetically distinct and had minimal recombination ( Figure 4B ) . The overall TMRCA was 52 months , while the TMRCAs for the distinct plasma and CSF populations were 6 and 8 months , respectively . Together , these results indicate that the mother likely transmitted two genetically distinct viruses to the child , and one variant was sequestered within the CNS while the other was maintained within the periphery . A multiple variant transmission event was observed in one additional compartmentalized subject ( Figure S2 ) , two intermediate subjects ( Figure S3 and S4 ) as well as one equilibrated subject ( Figure S5 ) . For the compartmentalized and intermediate subjects , one virus was sequestered within the CNS . For the equilibrated subject , both transmitted variants expanded within the blood and CSF . These data suggest that some variants can get selectively established in the CNS early after transmission . Macrophage tropism of HIV-1 is associated with the ability to infect cells expressing low levels of CD4 [11] , [12] , [13] , while R5 T cell-tropic viruses infect these cells very poorly and require high levels of CD4 to enter cells [18] , [19] . However , different preparations of macrophages vary significantly in their ability to be infected due to differing levels of CD4 in separate preparations of monocyte-derived macrophages ( MDM ) ( Joseph et al . , in preparation ) . To avoid this confounding variability , we have turned to a cell line that has regulatable levels of CD4 and CCR5 , i . e . 293-Affinofile cells [34] . Entry phenotype was assessed by measuring the ability of pseudotyped reporter viruses to enter cells expressing either high or low levels of CD4 . Viruses pseudotyped with Env proteins derived from virus in equilibrated subjects were only able to infect Affinofile cells with high CD4 surface expression and were considered R5 T cell-tropic ( Figure 5A ) . Viruses pseudotyped with Env proteins derived from virus in intermediate subjects were also only able to infect cells expressing high levels of CD4 which we infer defines R5 T cell tropism ( Figure 5B ) . A partially evolved entry phenotype was observed in subjects 4007 and 4013 , where CSF variants were able to infect cells with low CD4 at modest levels , potentially identifying a precursor population to the low CD4 entry phenotype . Examples of viruses with a low CD4 entry phenotype were observed in two compartmentalized subjects ( 4049 and 4058 ) ( Figure 5C ) . For both of these subjects , only the Env-pseudotyped viruses derived from the compartmentalized CSF population , not virus from the blood , were able to infect cells with low CD4 surface expression . These results indicate that subtype C HIV-1 viruses with a low CD4 entry phenotype can be detected in the CSF of children , but this is not a common occurrence within the first three years . Thus , we hypothesize that in most children replication in the CNS is sustained by growth in T cells , while in a subset ( 10–20% of children with CNS compartmentalized virus ) the virus evolves to replicate in cells with low CD4 surface expression , potentially macrophages and/or microglia .
Our study design involves cross-sectional sampling and thus has several limitations , especially with regard to inferring temporal relationship . However , there is a wide distribution of ages of the subjects within the enrollment criteria allowing us to compare between different age groups and to draw correlations based on age . Also , the virus carries the history of longitudinal evolution in its sequence , thus allowing us to infer dates of bottlenecks in the history of viral replication . While cross-sectional analyses are inherently limited , we have some basis for suggesting temporal relationships in the observed phenomenon . Based on our results we hypothesize four distinct states to describe the relationship between virus in the CSF/CNS and virus in the blood/periphery . The first state has no genetic evidence for HIV-1 replication in the CNS , wherein the only virus detected in the CSF is genetically similar to that in the blood and is typically present at 1% or lower of the level in the blood , possibly due to some import or spill pathway from the blood into the CSF . The second state occurs prior to 18 months of age and involves minor compartmentalization of the CSF viral population , suggesting some local replication in the CSF/CNS but not to a level where the viral load increases in the CSF . In 10–20% of these children there is evidence for the initial evolution of virus that can use lower levels of CD4 , potentially on a path to becoming macrophage-tropic . The third state occurs in about half of the children older than 18 months of age , and in this state the viral population in the CSF shows strong evidence of genetic compartmentalization , indicative of local replication and evolution within the CNS . This is also accompanied by a higher relative viral load in the CSF due to the local production of virus well above the low level that is imported from the periphery . In about 10–20% of the children with compartmentalized virus in the CSF , we identified variants that had evolved to use low levels of CD4 , which we presume indicates that the virus was now growing in macrophages and/or microglia within the CNS . The fourth state involves multiple variant transmissions from mother to infant of which one variant preferentially replicates in the CNS and another replicates in the periphery . As HIV-1 replication in the CNS can contribute to neurological disease , further research should determine whether the ability to detect different states of CSF viral populations within the CNS of young children could guide strategies to monitor and prevent neurodevelopmental disorders in HIV-infected children . Vertically transmitted viruses are often highly homogeneous , representing infection seeded by a single variant and characterized by low diversity [35] , [36] , [37] , [38] . We identified five infants who appeared to be infected with multiple variants , with the viral populations in the remaining infants having a phylogenetic age consistent with the age of the infant , which we assume indicates infection with a single variant . Surprisingly , in four of the five infants infected with multiple variants , one of the variants was largely sequestered in the CNS/CSF . For this to occur , either one of the transmitted variants had a selective tropism for the CNS , or infection of the CNS was a low probability event influenced by the chance introduction of a founder virus . Alternatively , since the CNS is a somewhat immune-privileged site , the absence of the sequestered virus in the periphery may be due to selection by maternal antibodies or the initial infant immune response in that compartment . All of the env genes tested from three of these subjects ( 3002 , 3017 , and 4002; see Figure 5 ) showed that the pseudotyped viruses required high levels of CD4 to enter cells , i . e . were R5 T cell-tropic . In our previous work we found multiple variants in approximately 30% of infants infected vertically [38] , which is not substantially different from the number found here . The sequestration of virus in the CNS shortly after transmission suggests that inferring the number of transmitted variants based on the complexity of virus in the blood may result in an underestimate of the frequency of transmission of multiple variants . We can infer several other features of compartmentalization by comparing the age of the infant to the inferred age of the viral population using BEAST for those infants infected with a single variant . In the remaining 10 infants with compartmentalization who were infected with a single variant , the age of the CSF/CNS compartmentalized viral population was significantly less than the age of the entire viral population when compared to the age of the infant ( P = 0 . 002; Wilcoxon signed rank test ) . In these cases it appears that compartmentalization is established after the initial stages of infection , with the compartmentalized virus emerging on average at 13 . 5 months but with outgrowth in the CNS only becoming apparent in the CSF approximately 18 months after birth . Thus we can identify two distinct pathways to compartmentalization: early sequestration of a transmitted virus in the CNS , or the later establishment of independently replicating virus that originates in the periphery . The compartmentalized lineage in the intermediate group appeared earlier after birth ( mean 5 . 1 months ) compared to the compartmentalized lineage in the compartmentalized group ( mean 13 . 5 months ) ( P = 0 . 008; Mann-Whitney test ) . This may be an indication that the intermediate state represents susceptibility to viral replication in the CNS but that there is a subsequent bottleneck that defines the CNS population . Genetically compartmentalized R5 T cell-tropic and macrophage-tropic HIV-1 subtype B populations have been shown to be associated with neurological complications in adults [9] . The macrophage-tropic populations were genetically diverse , representing established CNS infections , while the R5 T cell-tropic populations were clonally amplified and associated with pleocytosis [9] . Macrophage-tropic HIV-1 variants are generally characterized by the ability to infect cells with low CD4 surface expression [11] , [12] , [13] . However , infection using MDM from healthy donors is highly variable , and the variability is correlated with different levels of CD4 ( Joseph et al . , in preparation ) . For this reason , it is more quantitative to use a cell line where the levels of CD4 are regulatable and reproducible . Thus we have used Affinofile cells [34] as a surrogate for the entry phenotype of viruses able to use low levels of CD4 versus those requiring high levels of CD4 . Our results demonstrated that compartmentalized R5 T cell-tropic and what we infer to be macrophage-tropic populations can also be found in the CSF of children infected with HIV-1 subtype C . A partial entry phenotype was observed in two intermediate subjects; we hypothesize that this is evidence for the initial evolution of virus that can use lower levels of CD4 , potentially on a path to becoming macrophage-tropic . Viral replication in the CNS results in the local production of inflammatory and neuronal destruction molecules such as monocyte chemoattractant protein ( MCP-1 ) , neopterin , IP-10 , and neurofilament light subunit ( NFL ) . Production of these inflammatory markers has been observed in animal models [39] , [40] and has been linked to HIV-1-associated neurocognitive damage in adults [41] , [42] , [43] . The potential for long term neurocognitive damage in children as a result of HIV-1-associated production of inflammatory markers within the CNS , our findings of compartmentalized viral replication with viral lineage established at 13 . 5 months on average , and the ability of a transmitted variant to become sequestered in the CNS shortly after transmission adds further justification to the policy of early initiation of antiretroviral treatment in children , in this case as part of an effort to prevent the establishment of compartmentalized viral populations that may contribute to neurological complications .
The study was approved by the Institutional Review Boards of the University of North Carolina at Chapel Hill and the University of Malawi College of Medicine in Blantyre . Permission to participate in the research study was obtained for all children through written informed consent by the caregiver . All subjects included in this study were HIV-1 subtype C-infected children between 3 and 35 months of age . HIV-1 infection was verified at time of enrollment by a positive PCR for HIV DNA/RNA if <18 months of age or two positive rapid HIV antibody tests after age 18 months . Samples were collected at a one pre-HAART baseline visit for all subjects . CSF and blood plasma samples were used for viral genetic compartmentalization and env protein phenotypic analyses . Blood plasma and CSF HIV-1 viral loads ( copies/mL ) were determined by the UNC Chapel Hill Center for AIDS Research Virology Core . Subtype C HIV-1 RNA was isolated from blood plasma and CSF samples as previously described [16] . Briefly , viral RNA was isolated from blood plasma and CSF samples ( 140 µL ) using the QIAmp Viral RNA Mini kit ( Qiagen ) . Prior to RNA isolation , all blood plasma and CSF samples were pelleted ( 0 . 1–0 . 5 mL ) by centrifugation at 25 , 000×g for 1 . 5 hours at 4°C to increase template number and improve sampling . Purified viral RNA ( 10–50 µl ) was reverse transcribed using Superscript III Reverse transcriptase ( Invitrogen ) and an oligo-d ( T ) primer according to the manufacturer's instructions . Single genome amplification ( SGA ) of the full-length HIV-1 env gene through the 3′ LTR U3 end was conducted as previously described [30] . Briefly , cDNA was endpoint diluted and nested PCR was completed using Platinum Taq High Fidelity polymerase ( Invitrogen ) and the primers Vif1 [30] and 2 . R3 . B6R ( 5′-TGAAGCACTCAAGGCAAGCTTTATTGAGGC-3′; nt 9607 to 9636 ) , and EnvA [30] and Low2c ( 5′-TGAGGCTTAAGCAGTGGGTTCC-3′; nt 9591 to 9612 ) , were used for the first and second rounds of PCR , respectively . PCR amplicons were sequenced from the start of env through env gp41 , gp160 end ( HXB2 numbering of positions 6110–8833 ) . Chromatograms with double peaks , indicating amplification from more than one cDNA template , as well as sequences with frameshift mutations resulting in premature stop codons , were excluded from analysis . DNA sequences alignments of env genes were performed using ClustalW [44] . Sequences for each subject were codon aligned ( MEGA 4 . 0 ) and phylogenetic trees were generated using neighbor-joining method ( MEGA 4 . 0 ) [45] . Compartmentalization of viral sequences was assessed using the Slatkin-Maddison test [31] available through HyPhy [46] using 10 , 000 permutations . No contamination occurred between samples ( Figure S6 ) . A Bayesian Markov Chain Monte Carlo ( MCMC ) approach , as implemented in BEAST v . 1 . 6 . 1 [33] , estimated TMRCA for each patient sample . A substitution rate of 3 . 5×10−5 substitutions/site/generation was fixed under a strict clock model , as determined by calculation of inter-patient percent difference in the plasma nucleotide sequence . The HKY nucleotide substitution model had estimated base frequencies and a gamma-distributed rate heterogeneity ( 4 gamma categories ) . A coalescent Bayesian Skyline tree prior with a Piecewise-constant skyline model was used ( 4 groups ) . The MCMC algorithm was run for 30 million generations , logging every 1000 and with a 10% burn-in . The results from at least two independent runs were combined , and the effective sample size for all estimates was >200 . A generation time of 1 . 5 days was used for estimation of time to the MRCA [47] . SGA amplicons , selected based on the subject's phylogenetic tree structure , were re-amplified from the first-round nested PCR product using the Phusion hot start high-fidelity DNA polymerase ( Finnzymes ) and the primers EnvAClon ( 5′ CACCGGCTTAGGCATCTCCTGTGGCAGGAAGAA-3′; nt 5950–5982 ) and EnvN [30] following the manufacturer's instructions . HIV-1 env amplicons were then gel purified using the Qiagen gel extraction kit ( QIagen ) . Purified HIV-1 env genes ( 50 ng ) were cloned into the pcDNA3 . 1D/V5-His-TOPO expression vector ( invitrogen ) using the pcDNA 3 . 1 directional TOPO expression kit ( Invitrogen ) and the entire cloning reaction ( 6 µl ) was transformed into MAX Efficiency Stbl2 competent cells ( 50 µl ) as per the manufacturer's instructions . Bacterial colonies were screened for correct insertion of the HIV-1 env gene using colony PCR , and DNA was extracted from 3–6 positive colonies using the Qiaprep spin miniprep kit ( Qiagen ) . 293T cells were cultured in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) and 100 mg/ml of penicillin and streptomycin . 293-Affinofile cells [34] were maintained in DMEM supplemented with 10% dialyzed FBS ( 12–14 kD dialyzed; Atlanta Biologicals ) and 50 mg/ml blasticidin ( D10F/B ) . The Affinofile cell line was generously provided by Dr . Ben-Hur Lee . Each Env-pseudotyped luciferase reporter virus was generated using the Fugene 6 transfection reagent and protocol ( Roche ) to co-transfect 293T cells with an env expression vector and the pNL4-3 . LucR-E- HIV-1 backbone ( obtained from the NIH AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH ) . Prior to transfection , 293T cells were seeded at a density of 4 . 8×105 cells/well in 6-well tissue culture plates coated with 10% poly-L-lysine . Transfection medium was replaced five hours post-transfection with fresh culture medium and the cells were incubated at 37°C for 48 hours , after which viral supernatants were filtered with 0 . 45 µM filters ( Millipore ) and stored at −80°C . 293-Affinofile cell [34] CD4 and CCR5 receptor expression was induced with doxycycline ( doxy; Invitrogen ) and ponasterone A ( ponA; Invitrogen ) , respectively , and induced as previously described [16] . Briefly , cells were induced with doxy ( 0 ng/ml or 6 ng/ml ) and ponA ( 5 µM ) for 18–24 hours at 37°C and receptor expression was measured using quantitative fluorescence-activated cytometry ( qFACS ) following staining with either phycoerythin ( PE ) -conjugated anti-human CD4 antibody ( clone Q4120 , BD Biosciences ) or PE-conjugated mouse anti-human CCR5 antibody ( clone 2D7 , BD Biosciences ) . CD4 and CCR5 receptor levels were quantified using QuantiBRITE beads ( BD Biosciences ) . Env-pseudotyped luciferase reporter viruses were first titered in triplicate in a 96-well plate format on 293-Affinofile cells [34] expressing the maximum induction levels for both CD4 ( 6 ng/ml doxy ) and CCR5 ( 5 µM ponA ) surface expression as previously described [16] . In order to ensure that each infection assay was performed within the linear range , we used the volume of each virus needed to produce 800 , 000 relative light units ( RLUs ) when used to infect Affinofile cells expressing the highest levels of CD4 and CCR5 . Two days prior to infection , 96-well , black tissue culture plates were coated with 10% poly-L-lysine and then seeded with 293-Affinofile cells at a density of 1 . 85×104 cells/well . 18–24 hours later , expression of CD4 and CCR5 was induced at two conditions in triplicate: CD4high/CCR5high ( 6 ng/ml doxy and 5 µM ponA , respectively ) and CD4low/CCR5high ( 0 ng/ml doxy and 5 µM ponA ) . 18 to 24 hours later , the induction medium was removed and gently replaced with 100 µl of fresh , warmed culture medium containing env-pseudotyped virus . The infection plates were spinoculated [48] at 2 , 000 rpm for 2 hours at 37°C , and then incubated for an additional 48 hours at 37°C . Infection medium was then removed , the cells were lysed , and luciferase activity was assayed using the luciferase assay system ( Promega ) . The HIV-1 env nucleotide sequences determined in this study have been deposited in GenBank under accession numbers KC186127-KC187733 . | Genetically compartmentalized human immunodeficiency virus type 1 ( HIV-1 ) subtype B populations can be variably detected in the cerebrospinal fluid ( CSF ) of adults . Compartmentalization is indicative of local CNS replication , and late in disease is linked to HIV-associated dementia ( HAD ) . Compartmentalized viral populations can comprise either CCR5-using T cell-tropic or macrophage-tropic virus . Little is known about CNS infection in children or the ability of subtype C HIV-1 to evolve macrophage-tropic variants . We examined viral populations in the blood and CSF of HIV-1 subtype C-infected children . We found an intermediate level of compartmentalization in about half of the children under 18 months of age . About 50% of children older than 18 months had clearly compartmentalized virus in the CSF/CNS , and in some cases CSF virus evolved a low CD4 entry phenotype . In some of the children two variants were transmitted from the mother . In several of these cases one of the transmitted viruses was replicating in the CNS while the other was found predominantly in the blood/periphery . Our results suggest that compartmentalized CSF/CNS populations can be detected in up to 50% of children by year three , either established early in the infection or through sequestration of a transmitted variant within the CNS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"neuroinvasiveness",
"viral",
"transmission",
"and",
"infection",
"microbiology",
"immunodeficiency",
"viruses",
"neurovirulence",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"viral",
"entry",
"biology",
"pathogenesis",
"viral",
"evolution",
"virology"
] | 2012 | Central Nervous System Compartmentalization of HIV-1 Subtype C Variants Early and Late in Infection in Young Children |
Rift Valley fever virus ( RVFV ) ( genus Phlebovirus , family Bunyaviridae ) is an arbovirus that causes severe disease in humans and livestock in sub-Saharan African countries . Although the MP-12 strain of RVFV is a live attenuated vaccine candidate , neuroinvasiveness and neurovirulence of MP-12 in mice may be a concern when vaccinating certain individuals , especially those that are immunocompromised . We have developed a novel , single-cycle replicable MP-12 ( scMP-12 ) , which carries an L RNA , M RNA mutant encoding a mutant envelope protein lacking an endoplasmic reticulum retrieval signal and defective for membrane fusion function , and S RNA encoding N protein and green fluorescent protein . The scMP-12 underwent efficient amplification , then formed plaques and retained the introduced mutation after serial passages in a cell line stably expressing viral envelope proteins . However , inoculation of the scMP-12 into naïve cells resulted in a single round of viral replication , and production of low levels of noninfectious virus-like particles . Intracranial inoculation of scMP-12 into suckling mice did not cause clinical signs or death , a finding which demonstrated that the scMP-12 lacked neurovirulence . Mice immunized with a single dose of scMP-12 produced neutralizing antibodies , whose titers were higher than in mice immunized with replicon particles carrying L RNA and S RNA encoding N protein and green fluorescent protein . Moreover , 90% of the scMP-12-immunized mice were protected from wild-type RVFV challenge by efficiently suppressing viremia and replication of the challenge virus in the liver and the spleen . These data demonstrated that scMP-12 is a safe and immunogenic RVFV vaccine candidate .
Rift Valley fever virus ( RVFV ) , a member of the genus Phlebovirus within the family Bunyaviridae , carries a tripartite , single-stranded and negative–sense RNA genome [1]–[3] . The L RNA encodes the L protein , a viral RNA-dependent RNA polymerase; the M RNA encodes four proteins , including two accessory proteins , the NSm and 78-kDa proteins , and two major viral envelope proteins , Gn and Gc ( Gn/Gc ) ; and the S RNA uses an ambisense strategy to express the N protein and an accessory protein , NSs . In infected cells viral RNA synthesis occurs in the cytoplasm , while viral assembly and budding take place at the Golgi apparatus , where Gn/Gc accumulates . The virus is transmitted by mosquitoes and is maintained in nature , in sub-Saharan Africa , at least in part , by transovarial transmission . RVFV is able to infect various species of mosquitoes [4] and has the potential to spread to other areas of the world . Indeed , RVFV has already spread outside of the African continent to the Arabian Peninsula . The intentional spread of RVFV is also a serious national biosecurity concern . Human infection usually results in febrile illness , but may also cause viral hemorrhagic syndrome , encephalitis , and ocular disease [5]–[7] . RVFV also infects domestic ruminants and causes high mortality and spontaneous abortion rates with severe hepatic disease [8] . Introduction of RVFV to other areas of the world , including North and South America , Asia , and Europe , could cause serious public health problems and economic losses . RVFV spread can be prevented by the effective vaccination of animals and humans [1] . RVFV is considered to be serologically monotypic [9]–[11] , and humoral immunity , particularly neutralizing antibodies that recognize Gn/Gc , is important for protection [12]–[20] . Although a good human RVFV vaccine is urgently needed , there is no approved vaccine that can be adapted to massive vaccination programs . The MP-12 strain of RVFV [21] , which was developed by the serial passage of wild-type ( wt ) RVFV strain ZH548 in the presence of the mutagen 5-fluorouracil , is markedly attenuated and yet retains its immunogenicity [22]–[28]; hence , MP-12 is a promising live vaccine candidate for both human and veterinary use . However , intraperitoneal ( i . p . ) inoculation of young mice with MP-12 can result in efficient virus replication in the central nervous system ( CNS ) ( J . Morrill et al , unpublished data ) . Furthermore , i . p . inoculation of SCID mice with MP-12 results in the development of neurological signs and death of all mice [29] . These data suggest that MP-12 can invade the CNS and undergo efficient replication in immunocompromised animals , and may potentially do so in immunocompromised humans as well . However , neurovirulence tests in rhesus macaques show MP-12 to be less neuroinvasive and neurovirulent than acceptable lots of yellow fever or measles vaccine ( 28 ) . Even so , neuroinvasiveness and neurovirulence is of concern when considering RVFV immunization of the general public , given the diversity of ages , health statuses and genetic backgrounds . Thus , it is important to develop highly immunogenic RVFV vaccines with reduced or no neurovirulence . To develop a safe and immunogenic RVF vaccine , we have generated a novel , single-cycle replicable MP-12 ( scMP-12 ) , which does not cause systemic infection in immunized hosts , while resulting in expression of all viral structural proteins and production of noninfectious , virus-like particles ( VLPs ) in naïve cells infected with scMP-12 . The scMP-12 did not show any sign of neurovirulence after intracranial inoculation into suckling mice , demonstrating its safety . scMP-12-immunized mice elicited neutralizing antibodies and were efficiently protected from wt RVFV challenge by inhibiting wt RVFV replication in various organs and viremia . Our data suggest that scMP-12 has excellent potential to be developed as a safe RVF vaccine .
All mouse studies were performed in facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care in accordance with the Animal Welfare Act , NIH guidelines and U . S . federal law . The animal protocol was approved by the UTMB Institutional Animal Care and Use Committee . The wt RVFV ZH501 strain was used in an enhanced ABSL-3 laboratory within the Galveston National Laboratory at UTMB in accordance with NIH guidelines and U . S . federal law . Vero E6 cells and BSR-T7/5 cells [30] , the latter of which stably express T7 RNA polymerase , were maintained as described previously [31] , [32] . BHK-21 cells were maintained in minimal essential medium ( MEM ) α medium ( Gibco ) supplemented with 5% fetal bovine serum ( FBS ) . The MP-12 strain of RVFV was generated by reverse genetics [31] . A standard PCR-based method , in which pProT7-M encoding antiviral-sense M RNA [31] served as a template , was used to generate pProT7-M-Gn/GcΔ5 , which expresses M-Gn/GcΔ5 RNA carrying a deletion between nucleotide positions 3597 and 3611 in the M segment . A Quickchange II site-directed mutagenesis kit ( Agilent Technologies ) was used to obtain pProT7-M-Gn/GcΔ5-derived mutants , each of which carried an amino acid substitution ( s ) within a putative fusion peptide . Plasmid pCAGGS-bla-G was constructed by inserting the Not I-EcoR V fragment of pCX4-bsr [33] , which contains the encephalomyocarditis virus internal ribosomal entry site and blasticidin-resistant gene , into the Not I and Stu I sites of pCAGGS-G , which carries the entire open reading frame ( ORF ) of MP-12 M RNA encoding 78-KDa , NSm , Gn and Gc proteins . The sequences of all of the constructs were confirmed not to contain unwanted mutations . MP-12 , scMP-12 , and MP-12-based , 2-segmented virus replicon particles ( VRP ) were generated by using a reverse genetics system [31] . Briefly , BSR-T7/5 cells were co-transfected with plasmids encoding the L , N , and Gn/Gc proteins , and anti-viral sense L , M , and S RNAs for MP-12 recovery . scMP-12 recovery was performed by using a similar method with the following modifications: a plasmid expressing the S RNA carrying an N gene and green fluorescent protein ( GFP ) ( S-GFP RNA ) was used in place of that expressing the S RNA; a plasmid encoding M-Gn/GcΔ5 RNA with two amino-acid substitutions , F826N and N827A , was used in place of that expressing the M RNA; and a plasmid expressing the MP-12 Gn/Gc optimized for bovine codon usage was used in place of that expressing the MP-12 Gn/Gc to prevent or minimize homologous RNA recombination events between expressed mRNA encoding Gn/Gc and the replicating M RNA mutant . For VRP recovery , a plasmid expressing S-GFP RNA was used in place of the plasmid encoding the S RNA and the plasmid encoding the M RNA was eliminated . Culture fluid was collected at 5 , 10 and 10 days post transfection for MP-12 , scMP-12 , and VRP , respectively . Vero E6 cells were transfected with pCAGGS-bla-G , and incubated in the presence of 20 µg/ml of blasticidin from 1 day post-transfection . After obtaining blasticidin-resistant cell clones by limiting dilution , each cell clone was tested for Gn protein expression by indirect immunofluorescence with an anti-Gn monoclonal antibody ( R1-4D4 ) [34] , and a cell clone expressing highest levels of Gn was selected and designated as Vero-G cells . A standard plaque assay was used to determine the infectivity of MP-12 [31] . For determining the infectivity of scMP-12 and VRP , Vero-G cells in 6-well plates were inoculated with 400 µl of serially diluted samples and incubated for 1 h at 37°C . After removal of the inocula , cells were incubated with MEM containing 0 . 6% Tragacanth gum ( MP Biomedicals ) , 5% FBS , and 5% tryptose phosphate broth at 37°C . After 3 days incubation , cells were washed with phosphate-buffered saline ( PBS ) and fixed with PBS containing 4% paraformaldehyde for 20 min at room temperature . After removing paraformaldehyde and overlays , the cells were permeabilized with 0 . 1% Triton-X100 and incubated with anti-N rabbit polyclonal antibody , which was generated by injecting a purified , bacterially-expressed fusion protein consisting of glutathione-S-transferase and full-length MP-12 N protein into rabbits , followed by incubation with horseradish peroxidase-conjugated , anti-rabbit IgG antibody . The plaques were visualized with Nova RED peroxidase substrate ( Vector Laboratories , Burlingame , CA ) . This modified plaque assay was also used for observing plaque morphologies of MP-12 in Vero-G cells . The cell fusion assay was performed as previously described [35] , [36] with some modifications . Briefly , BSR-T7/5 cells were co-transfected with plasmids encoding the Venus , N , and L proteins , and M-Gn/GcΔ5 RNA or M-Gn/GcΔ5 RNA with single amino acid substitutions , and incubated at 37°C for 24 h . To initiate cell fusion , the cells were washed with Mg2+- and Ca2+-containing acidic PBS ( pH adjusted to 5 . 2 with citric acid ) and treated with the acidic PBS for 5 min . , and then incubated in complete medium at 37°C for 60 min . GFP signals in the cells were observed under a fluorescence microscope ( Zeiss ) . BSR-T7/5 cells were co-transfected with plasmids encoding the N and L proteins , and M-Gn/GcΔ5 RNA or its mutant . Twenty-four hours after transfection , cells were fixed with 4% paraformaldehyde and permeabilized with 0 . 1% Triton-X100 , or not permeabilized . Cells were incubated with the primary monoclonal antibody that recognizes Gn ( R1-4D4 ) or Gc ( R1-5G2 ) [37] for 1 h at room temperature and with the Alexa-594-conjugated secondary antibody for 1 h at room temperature , and observed under a fluorescence microscope . Cells were harvested by using a cell scraper and washed with PBS . After incubation of the harvested cells on ice for 20 min in cell lysis buffer ( 20 mM Tris-HCl , 150 mM NaCl , 1% Triton X-100 ) , the cell lysate was centrifuged at 2 , 000 rpm for 3 min by using a microcentrifuge . The resultant supernatant was mixed with the same amount of 2× sample buffer and boiled for 5 min . Equal amounts of samples were subjected to SDS-polyacrylamide gel electrophoresis . Proteins were electroblotted onto polyvinylidene difluoride membranes ( Millipore ) . After blocking the membrane with 1% bovine serum albumin for 1 h , the membranes were incubated with the primary antibody for 1 h at room temperature . After incubation with the secondary antibody for 1 h at room temperature , the blots were developed by using an ECL kit ( GE Healthcare ) . Anti-MP-12 mouse polyclonal antibody [31] was used to detect the virus-specific proteins . Total RNAs were extracted by using TRIzol reagent ( Invitrogen ) and subjected to Northern blot analysis as described previously [38] . Viral-sense-specific , digoxigenin-labeled RNA probes [31] and a digoxigenin system ( Roche ) were used for the detection of viral RNAs . The L RNA probe hybridizes with viral-sense L RNA at nucleotide positions 19–756 , the M RNA probe at nucleotide positions 1297–2102 , and the S RNA probe at nucleotide positions 39–776 from the 3′ ends of the viral-sense RNA segments . The probe that hybridizes with anti-viral sense S RNA binds at nucleotide positions 39–776 from the 5′ end of the anti-viral-sense S RNA segment . Culture medium harvested from plasmid-transfected cells or scMP-12-infected cells was clarified by centrifugation at 3 , 000 rpm for 15 min by using a tabletop centrifuge . The clarified supernatant was layered on top of a step sucrose gradient consisting of 20 , 30 , 50 , and 60% sucrose ( wt/vol ) and centrifuged for 3 h at 26 , 000 rpm at 4°C using a Beckman SW28 rotor [32] . The particles at the interface of 30 and 50% sucrose were collected , diluted and subjected to a second sucrose gradient centrifugation consisting of 20 , 30 , 50 , and 60% sucrose for 18 h at 4°C . The particles at the interface of 30 and 50% sucrose were collected and pelleted down through a 20% sucrose cushion at 38 , 000 rpm for 2 h at 4°C using a Beckman SW41 rotor . scMP-12 was serially passaged 10 times in Vero-G cells under the following three conditions for each passage: inoculation without sample dilution and harvest at 4 days p . i . , inoculation after 10 times sample dilution and harvest at 5–6 days p . i . , and inoculation after 100 times sample dilution and harvest at 7 days p . i . We visually inspected for an increase in the number of GFP-positive cells every day . Each of the culture fluids collected was also inoculated into Vero E6 cells , and the GFP signal was examined daily up to 5 days p . i . Two-day-old CD1 mice were intracranially inoculated with 104 PFU of MP-12 , scMP-12 , or the same volume of Hank's balanced salt solution ( HBSS ) . We monitored the mice for survival for 21 days . CD1 mice ( 5-week-old females ) were intramuscularly immunized with 104 PFU of MP-12 , 105 PFU of MP-12 , 105 PFU of scMP-12 , or 105 PFU of VRP . Thirty-six days later , blood was collected from the retro-orbital venous plexus of the mice . Forty days after vaccination , the immunized mice were challenged subcutaneously with 103 PFU of the virulent RVFV strain ZH501 , which was equivalent to approximately 1 , 000 times the 50% minimal lethal dose ( LD50 ) . The animals were observed for survival and clinical signs of disease for 21 days post-challenge . To determine the effect of immunization on virus replication , sera and specimens of liver , spleen and brain were harvested from randomly selected animals at 3 , 6 , 9 and 11 days post-challenge . Sera and 10% tissue homogenates were tested for virus presence and titer in Vero E6 cells , as previously described ( 39 ) . Serum neutralizing antibody titers were determined by using an 80% plaque-reduction neutralization test ( PRNT80 ) , as previously described [39] .
We designed the scMP-12 system as shown in Fig . 1A . scMP-12 carries a membrane-fusion defective mutant of Gn/Gc and is rescued by using a modified MP-12 reverse genetics system [31] , in which BSR-T7/5 cells stably expressing T7 polymerase [30] are co-transfected with three RNA-expression plasmids expressing the L RNA , a mutant M RNA encoding a membrane-fusion defective mutant of Gn/Gc , and a S-GFP RNA encoding the N and GFP proteins , as well as three protein expression plasmids encoding the L , N , and Gn/Gc proteins . The scMP-12 that is produced is infectious due to the presence of Gn/Gc and undergoes amplification in Vero-G cells stably expressing Gn/Gc . Inoculation of the amplified scMP-12 into naïve cells results in viral RNA synthesis , expression of viral proteins , including L , N and the fusion-defective Gn/Gc , and production of noninfectious VLPs containing the fusion-defective Gn/Gc . In immunized hosts , scMP-12 undergoes single cycle replication in infected cells , resulting in the intracellular accumulation of all of the viral structural proteins and the production of noninfectious VLPs; scMP-12 particles in the inoculum , viral proteins accumulated in scMP-12-infected cells and released noninfectious VLPs all serve as immunogens to elicit immune responses to RVFV proteins . Due to its characteristic single-cycle replication , it is highly unlikely that the scMP-12 can cause systemic infection or invade the CNS of immunized animals or humans . To isolate fusion-defective Gn/Gc mutants suitable for scMP-12 , we first developed a cell-to-cell membrane fusion assay . Phlebovirus glycoprotein-induced , virus-cell membrane fusion requires a low pH ( ∼pH 5 . 4 ) environment [35] . Exposure of cells expressing RVFV Gn/Gc to low pH conditions does not induce cell-to-cell fusion due to the absence of Gn/Gc at the plasma membrane; RVFV Gn/Gc accumulates at the Golgi apparatus and the endoplasmic reticulum ( ER ) in infected cells and in expressed cells . Phleboviruses have an ER retrieval signal of ∼5 amino acids in the cytoplasmic tail of Gc [40] , and removal of this signal in the Gc of Uukuniemi virus ( a Phlebovirus ) results in an accumulation of expressed Gn/Gc at the Golgi apparatus and plasma membrane [40] . Likewise , mutant MP-12 Gn/Gc lacking the terminal C-terminal 5-amino-acid residues of the Gc ( Gn/GcΔ5 ) primarily accumulated at the Golgi apparatus when expressed , and some mutant glycoprotein was translocated to the plasma membrane ( Fig . 2A ) . Exposure of the cells expressing Gn/GcΔ5 , but not those expressing wt Gn/Gc , to low pH conditions induced cell-to-cell membrane fusion; fusion was not observed at neutral pH conditions for cells expressing Gn/GcΔ5 ( Fig . 2B ) . These data suggest that Gn/GcΔ5 protein that localized to the plasma membrane was fusion-competent only under low pH conditions . We sought to generate fusion-defective Gn/Gc mutants by altering amino acids in the putative fusion peptide , which was previously predicted by computational studies and structural analysis [41] , [42] . Alignment of the predicted fusion peptide sequences of several Phleboviruses revealed the presence of a highly conserved cysteine residue at position 825 ( C825 ) , which is involved in a disulfide bond in the Gc [41] , and a phenylalanine residue at position 826 ( F826 ) ( Fig . 2C ) . Because hydrophobic residues are important for the insertion of fusion peptides into the cell plasma membrane [43] , we tested the fusion competence of a series of Gn/GcΔ5-derived mutants , in which the F826 was changed to a hydrophilic residue , or its surrounding hydrophobic residues and hydrophilic residues were changed to hydrophilic residues or hydrophobic residues , respectively ( Fig . 2D ) . While the V828N and P830N mutants retained fusion activity , the other mutants lost such activity ( Fig . 2D ) . Anti-Gn monoclonal antibody recognized all of the Gn/Gc mutants , while the anti-Gc monoclonal antibody R1-5G2 failed to detect the W821N and C825A mutants , implying an alteration of the Gc conformation occurred from these mutations . From the C823A , F826N and N827A mutants , all of which lost fusion activity and were detected by R1-5G2 , we selected F826N and N827A mutants for subsequent studies . Because development of scMP-12 is aimed at improving RVF vaccine safety , it is important to prevent the generation of infectious viruses in scMP-12-immunized hosts , as well as during scMP-12 preparation in cell culture . Hence , we tested several M RNA mutants , each encoding Gn/GcΔ5 , with different combinations of fusion peptide mutations and chose an M RNA mutant encoding Gn/GcΔ5 with the F826N and N827A mutations ( scMP-12 M RNA ) ( Fig . 1B ) for scMP-12 preparation primarily due to its excellent genetic stability . BSR-T7/5 cells were co-transfected with three protein expression plasmids expressing the L , N , and Gn/Gc proteins , and three RNA expression plasmids encoding the L , scMP-12 M , and S-GFP RNAs . The GFP signal generated in scMP-12-infected cells facilitated the monitoring of scMP-12 replication . We also generated a VRP , an MP-12-based virus replicon particle ( VRP ) carrying only the L and S-GFP RNAs . Because other groups have reported the generation of a VRP ( also called RVFV replicon particles ) carrying the L and S-GFP RNAs derived from wt RVFV [44] , [45] , we refer to the wt virus-based VRP as VRPwt to distinguish between it and the MP-12-based VRP used in this study . MP-12 was rescued as previously described [31] , and used as a positive control . Culture fluids from MP-12 samples were collected at 5 days post-transfection , while those from the scMP-12 and VRP samples were collected at 10 days post-transfection; these samples were defined as P0 samples . To amplify and titrate the scMP-12 samples , we generated Vero-G cells stably expressing MP-12 Gn/Gc , and found the expression levels of Gn/Gc in Vero-G cells to be roughly one-fourth of the levels for MP-12-infected Vero cells at 12 h post-inoculation ( p . i . ) ( Fig . 3A ) . Like MP-12-infected Vero E6 cells , Gn and Gc signals primarily accumulated in perinuclear regions of Vero-G cells ( Fig . 3B ) . We independently inoculated the P0 samples of scMP-12 and VRP into Vero-G cells and obtained passage 1 ( P1 ) samples after 10 days p . i . These P1 samples were predominantly used for subsequent studies . MP-12 , scMP-12 and VRP formed large , medium and small plaques , respectively , in Vero-G cells , in which plaques were visualized by anti-N protein antibodies ( Fig . 4A ) . Inoculation of MP-12 , scMP-12 or VRP into Vero-G cells at a multiplicity of infection ( MOI ) of 0 . 05 showed efficient MP-12 replication with maximum titers ∼108 PFU/ml at 3 days p . i . ( Fig . 4B ) . scMP-12 replicated to ∼106 PFU/ml at 2–3 days post-infection , whereas the titers of the VRP were roughly 5–10 times lower than those of scMP-12 ( Fig . 4B ) . As expected , we observed efficient accumulation of the three viral RNA segments in Vero-G cells infected with MP-12 or scMP-12 , and L and S-GFP RNAs in VRP-infected Vero-G cells ( Fig . 4C ) . We purified the particles produced from Vero-G cells infected with scMP-12 , MP-12 or VRP by sucrose gradient centrifugation . Western blot analysis of purified particles using anti-MP-12 antibody showed the presence of Gn/Gc and N proteins in all samples ( Fig . 4D ) . The origin is unknown for two bands found , one that migrated more slowly and the other faster than the Gn/Gc of the MP-12 sample in the gel . Northern blot analysis of viral RNAs extracted from the purified particles showed packaging of three viral RNAs in MP-12 and scMP-12 samples and that of the L and S-GFP RNAs in the VRP sample ( Fig . 4D ) ; the abundance of each of the viral RNAs was roughly proportional to the titers of MP-12 , scMP-12 and VRP at day 3 ( Fig . 4B ) . These data show that the scMP-12 underwent efficient replication and amplification in Vero-G cells . To examine scMP-12 replication in naïve cells , we inoculated MP-12 , scMP-12 or VRP into naïve BHK cells and examined the accumulation of viral proteins and RNAs ( Fig . 5A ) . Efficient accumulation of the Gn/Gc and N proteins occurred in MP-12-infected cells . Accumulation of the N and Gn/Gc proteins also occurred in scMP-12-infected cells , with lower levels of Gn/Gc accumulation as compared to MP-12-infected cells . VRP-inoculated cells showed an accumulation of the N protein , but not the Gn/Gc protein . Northern blot analysis showed that the three viral RNAs replicated in MP-12-infected cells and in scMP-12-infected cells , and L and S-GFP RNAs replicated in VRP-infected cells . An RNA probe that specifically binds to anti-viral-sense S RNA clearly demonstrated N mRNA synthesis in these RNA samples ( Fig . 5A , right panels ) . Thus , the scMP-12 underwent efficient viral RNA synthesis and viral protein accumulation in infected naïve cells . We next purified the particles released from scMP-12-infected BHK cells by sucrose gradient centrifugation and detected viral proteins in the purified particles ( Fig . 5B ) . The purified particles produced from MP-12-infected cells and VRP-infected cells served as a positive control and a negative control , respectively . Western blot analysis showed the production of MP-12 particles in the positive control by demonstrating the N and Gn/Gc proteins . No Gn/Gc signal was detected in the VRP sample , whereas the scMP-12 sample showed a low level of Gn/Gc signal . Both scMP-12 and VRP samples showed low levels of the N protein signal . Because synthesis of the Gn/Gc proteins did not occur in VRP-infected cells , it is highly unlikely that the N protein in the VRP sample represents released VRP . Continuous sucrose gradient centrifugation of culture fluid of MP-12-infected cells showed sedimentation of N protein with the purified virions as well as to lower sucrose density fractions [46] , suggesting the release of N protein which is not associated with virus particles from infected cells . Furthermore , release of N protein not associated with viral envelope proteins was reported in studies of RVFV VRP [45] and Crimean-Congo hemorrhagic fever virus [47] . Hence , the N protein signal in the VRP sample most probably represents the N protein that was not associated with virus particles . Likewise , most of the N signal in the scMP-12 sample was probably derived from the non-VLP-associated N protein . Nonetheless , the Gn/Gc signal in the scMP-12 sample suggests the occurrence of low levels of VLP production from scMP-12-infected naïve cells . Inoculation of supernatant from MP-12-infected BHK cells , but not from scMP-12-infected BHK cells or VRP-infected BHK cells , into fresh BHK cells resulted in viral RNA synthesis ( Fig . 5C ) , demonstrating that the VLP produced from scMP-12-infected naïve cells was not infectious . To evaluate the genetic stability of scMP-12 , we performed 10 serial passages of scMP-12 in Vero-G cells under three different conditions , as described in Materials and Methods , and tested the generation of infectious viruses that undergo multiple cycles of replication in naïve cells . Multiple cycles of the scMP-12 amplification in Vero-G cells resulted in an increase in the numbers of GFP-positive cells during incubation in each passage , whereas an increase in the numbers of GFP-positive cells did not occur after inoculation of any of the passage samples in Vero cells , suggesting the absence of infectious viruses in all of the passaged samples . Also plaque assays using Vero cells did not show the presence of infectious viruses in any of the samples . Sequence analysis of the PCR products of scMP-12 M RNA showed that scMP-12 retained the introduced mutations after 10 passages under the three different conditions . These results demonstrate that scMP-12 stably retained the introduced mutations . We tested the neurovirulence of scMP-12 by intracranially inoculating 1 . 0×104 PFU of scMP-12 into 2-day-old CD1 mice and monitoring for survival and clinical signs for 21 days p . i . As controls , HBSS and the same titer of MP-12 were inoculated . All MP-12 infected mice died by 3 days p . i . , whereas all mice inoculated with scMP-12 or HBSS survived and did not show any clinical signs of disease ( Fig . 6 ) , demonstrating the absence of detectable levels of neurovirulence in scMP-12 . We intramuscularly inoculated 5-week-old female CD1 mice once with 105 PFU of scMP-12 and determined the PRNT80 titers at 36 days p . i . As controls , mice were inoculated with 105 PFU of VRP , 105 PFU of MP-12 , 104 PFU of MP-12 , or HBSS ( Fig . 7 ) . HBSS-inoculated mice had no detectable neutralizing antibody titers , while mice inoculated with 105 PFU of MP-12 and 104 PFU of MP-12 had a mean PRNT80 titer of 1∶1 , 477 and 1∶310 , respectively . The mean PRNT80 titers of the mice immunized with 105 PFU of scMP-12 and 105 PFU of VRP were 1∶238 and 1∶38 , respectively; the difference in the PRNT80 titers was statistically significant . Thus , the mice immunized with 105 PFU of scMP-12 elicited neutralizing antibody titers that were statistically higher than those immunized with 105 PFU of VRP and were comparable to those immunized with 104 PFU of MP-12 . We next tested whether scMP-12 immunization protects mice from wild-type RVFV challenge . Five-week-old female CD1 mice were intramuscularly inoculated once with 105 PFU of scMP-12 , 105 PFU of VRP , 105 PFU of MP-12 , 104 PFU of MP-12 , or HBSS . At 40 days post-immunization , the mice were challenged subcutaneously with 1 . 0×103 PFU of the ZH501 strain of RVFV and their survival was monitored for 21 days p . i . ( Fig . 8 ) . All HBSS-inoculated mice died by 10 days p . i . , whereas all mice immunized with 105 PFU of MP-12 survived . Most of the mice immunized with 104 PFU of MP-12 or 105 PFU of scMP-12 survived , yet 1 of the 19 MP-12-immunized mice died at day 11 , and 3 of the 29 scMP-12-immunized mice died , one at day 10 and two at day 19 , respectively . In contrast , 45% of the VRP-immunized mice died by day 12 p . i . , demonstrating that scMP-12 immunization protected most of the mice from wt RVFV challenge , and scMP-12-induced protection was better than the VRP-induced protection . To study the extent to which scMP-12-induced immune responses suppressed wt virus replication upon challenge , HBSS-inoculated mice and mice immunized once with 105 PFU scMP-12 , 105 PFU VRP , 105 PFU MP-12 , or 104 PFU MP-12 were challenged with the ZH501 strain of RVFV , as described above , and the virus titers in serum , liver , spleen and brain were determined at days 3 , 6 , 9 and 11 post-challenge ( Fig . 9 ) . At day 3 p . i . , 4 out 5 HBSS-inoculated mice had >105 PFU/ml of viremia , and one and three mice showed virus replication in the liver and the spleen , respectively . Efficient virus replication in the brain also occurred in HBSS-inoculated mice from days 5 to 9 p . i . In contrast , mice immunized with VRP or MP-12 showed neither viremia nor virus replication in the liver , spleen or brain . scMP-12 immunization also prevented viremia and virus replication in the liver and spleen , while two mice , one having no detectable PRNT80 titer and the other having a PRNT80 titer of 1∶20 , showed virus replication in the brain at day 9 p . i .
By using a membrane fusion assay and newly established Vero-G cells , we generated scMP-12 and tested its potential as a safe and immunogenic RVFV vaccine . scMP-12 amplified efficiently in Vero-G cells and stably retained the introduced mutations in ten serial passages in this cell line under three different experimental conditions . In infected naïve cells , scMP-12 underwent efficient viral RNA synthesis and accumulated viral proteins , including Gn/Gc , and produced low levels of non-infectious VLPs . The scMP-12 did not show any sign of neurovirulence after intracranial inoculation into 2-day-old mice , demonstrating excellent safety . scMP-12 immunization in mice induced neutralizing antibodies , whose titers were higher than those in VRP-immunized mice , and protected most of them from wt RVFV challenge by suppressing viremia and wt RVFV replication in the liver and the spleen . Taken together , we consider that scMP-12 has an excellent potential to be developed as a novel safe RVFV vaccine . We examined effects of mutations within the putative fusion peptide for membrane fusion ( Fig . 2 ) . The crystal structure of RVFV Gc suggested that V828 and hydrophobic residues W821 and F826 within the putative fusion peptide serve as a membrane anchor during the pre-fusion step [41] . By substituting the hydrophobic residues for hydrophilic residues in the putative fusion peptide , we experimentally demonstrated that an F826N mutation , but not V828N mutation , abolished membrane fusion . Anti-Gc monoclonal antibody did not recognize Gc carrying W821N , and possibly a van der Waals interaction between W821 and F826 was disrupted by this mutation , leading to Gc structural alteration [41] . C825 is highly conserved among Phleboviruses ( Fig . 2C ) , and the C825A mutant was defective for the fusion function . Because C825 is involved in a disulfide bond in Gc [41] , and the anti-Gc monoclonal antibody did not recognize the C825A mutant , a lack of fusion function in this mutant was probably due to the structural alteration of Gc . Because other Phleboviruses also encode the ER retrieval signal in the Gc cytoplasmic tail , development of similar membrane fusion assays for other Phleboviruses would be possible . Recently , others also reported the utility of the RVFV membrane fusion assay that uses a plasmid transfection method [48] . Experiments using such fusion assays , which employ a conventional plasmid transfection , will be valuable for further understanding of the membrane fusion mechanism in Phleboviruses , and identification and evaluation of antivirals that suppress viral membrane fusion activity [48] . The data that RVFV spread can be prevented by effective vaccination of animals and humans [1] and that neutralizing antibodies , the majority of which recognize Gn/Gc protein , play a critical role in protection [12]–[20] led to development of several different types of RVFV vaccine candidates that primarily aim to elicit high titers of neutralizing antibodies . Formalin-inactivated RVFV vaccine requires several immunizations to induce and maintain protective immunity [49] , [50] . In contrast , several attenuated RVFV mutants , including MP-12 , MP-12-derived mutants carrying a modified cellular gene in place of the NSs gene [51] , and a wt RVFV-derived avirulent mutant lacking NSs and NSm genes , both of which are viral virulence factors [52]–[55] , demonstrated excellent protective immunogenicity against wt RVFV after a single immunization of animals [56] . Examples of other vaccine candidates are VLPs [57]–[59] , recombinant vaccinia viruses encoding Gn and Gc proteins [29] , alphavirus encoding the Gn protein [60] , alphavirus replicon encoding the Gn protein [61] , and a soluble ectodomain of the Gn protein [62] . Most of these vaccine candidates have used multiple dose immunization protocols to confer complete protection to immunized rodents against wt RVFV challenge . Single immunization of mice with scMP-12 ( Fig . 8 ) , VLP expressing low levels of viral N protein in infected cells [59] or VRPwt expressing both L and N proteins in inoculated cells [44] , [63] showed good protection of the immunized mice from lethal wt RVFV challenge . This finding may imply that the expression of the N protein and probably also the L protein in immunized animals facilitated development of strong protective immune responses . In addition , viral-replicating , single-stranded RNA and the incoming RNA virus nucleocapsids activate the innate immune system through interaction with the host pattern recognition receptor , e . g . RIG-I [64]–[69] , and potentiates the adaptive immune responses [70] . Moreover , viral RNAs in virus particles have an adjuvant effect for augmenting host-adaptive immune responses through a Toll-like receptor 7 signaling pathway in dendritic cells [71] , [72] . Therefore , it is likely that incoming nucleocapsids of scMP-12 , intracellular viral RNAs accumulated in scMP-12-infected naïve cells , and viral RNAs in the released VLPs all contributed to enhancement of the host immune response , making scMP-12 highly immunogenic . Importantly , scMP-12 was more immunogenic than VRP ( Fig . 7 ) , and the scMP-12-immunized mice were protected from wt RVFV challenge more efficiently than the VRP-immunized mice ( Fig . 8 ) ; hence , the expression of Gn/Gc in cells supporting scMP-12 replication and viral RNA containing VLPs produced by cells in which scMP-12 replicated augmented the protective immune response . A lack of neurovirulence and the characteristic single-cycle replication property of scMP-12 demonstrate that scMP-12 is superior to MP-12 in safety , as MP-12 killed all of the 2-day-old mice following intracranial inoculation , whereas scMP-12 was less immunogenic than MP-12; neutralizing antibody titers in mice immunized with 105 PFU of scMP-12 were comparable to those immunized with 104 PFU of MP-12 and lower than those immunized with 105 PFU of MP-12 . Improvement of scMP-12 immunogenicity may be possible by generating a scMP-12 variant that produces a high abundance of VLPs following scMP-12 replication . Because substitution of several histidines in RVFV Gc with alanine inhibits membrane fusion activity but does not interfere with virion assembly [73] , the efficient production of noninfectious VLPs may occur in cells supporting replication of scMP-12 variants carrying some of these mutations . The finding of efficient scMP-12 amplification in Vero-G cells suggests that a scMP-12-based vaccine stock can be prepared in Vero-G cells or their equivalent without plasmid transfection , thereby allowing the production costs of the scMP-12-based vaccine to be comparable to those for MP-12 . scMP-12 and VRP produced plaques in Vero-G cells , showing the utility of Vero-G cells for easy titration and characterization of RVFV mutants lacking functional Gn/Gc proteins . We noted that MP-12 replicated roughly 10 times better in Vero-G cells than in Vero E6 cells ( Fig . 4B and [31] ) , which led us to suggest that higher levels of intracellular Gn/Gc accumulation augments MP-12 production . Likewise , an increase in the abundance of intracellular Gn/Gc in scMP-12-replicating cells may also enhance scMP-12 titers . Hence , the development of another Vero cell clone , in which expression levels of Gn/Gc are comparable to those in MP-12-infected Vero cells , would contribute to mass immunization programs using an scMP-12-based vaccine . The absence of infectious virus after 10 serial passages of scMP-12 in Vero-G cells under three different conditions demonstrated that homologous RNA recombination that can eliminate the mutations in scMP-12 M RNA did not occur between replicating scMP-12 M RNA and expressed mRNA encoding Gn/Gc in Vero-G cells , further indicating the utility and safety of Vero-G cells for preparation of the scMP-12-based vaccine . Lastly , we found that scMP-12 replicated ∼10 times better than did the VRP in Vero-G cells ( Fig . 4 ) . These data were consistent with the notion that M RNA serves important roles in viral RNA co-packaging [32] . Expression of GFP from the S-GFP RNA of scMP-12 facilitated easy monitoring of scMP-12 replication and generation of infectious viruses in scMP-12 preparations . VRPwt also used S-GFP-type RNA for easy monitoring of VRPwt replication [44] , [45] . However , vaccines encoding a foreign reporter gene , such as GFP , may not be appropriate for human use . Therefore , before we can develop a scMP-12-based human vaccine , it is necessary to test the replication competence , safety , and immunogenicity of scMP-12-based vaccine candidates lacking the NSs gene or of those carrying RVFV Clone 13-type S RNA lacking ∼70% of the NSs gene [74] . Our study was primarily aimed at the development of a safe and immunogenic human RVF vaccine , yet scMP-12 may be further developed as a veterinary vaccine . Others have reported that MP-12 is teratogenic in some cases [75] . Considering that scMP-12 only undergoes a single cycle of replication , it is unlikely cause disease in immunized animals . Vaccines that are compatible with a differentiation of infected and vaccinated animals ( DIVA ) are suitable for use as animal vaccines . Examples of replication-competent RVF DIVA vaccine candidates are RVFV Clone 13 lacking ∼70% of the NSs gene [74] , MP-12 lacking NSm , which elicited high titers of neutralizing antibodies in sheep and calves [76] , and wt RVFV-derived mutant virus lacking NSm and NSs , which induced protective immunity in immunized sheep [56] . The data that scMP-12 , which lacks an NSs gene , protected immunized mice from wt RVFV challenge ( Fig . 8 ) and that VRPwt , which also lacks an NSs gene , can induce protective immunity in sheep [63] indicate a potential for a scMP-12-based DIVA vaccine to reduce the incidence of RVF among humans and animals and to control this important pathogen [8] . | Rift Valley fever virus ( RVFV ) is a mosquito-borne zoonotic pathogen , which causes febrile illness , encephalitis and fatal hemorrhagic fever in humans and severe hepatic disease with high mortality and spontaneous abortion rates in ruminants . RVFV is endemic to the African continent . Because many different mosquito species support RVFV replication , the virus has the potential to spread to other areas of the world , such as North and South America , Asia , and Europe and could cause serious public health problems and economic losses . Consistent with this concern , RVFV has caused epidemic disease in the Arabian Peninsula . Currently , there is no approved vaccine suitable for mass vaccination programs of humans . Although the MP-12 strain of RVFV is a live attenuated vaccine candidate , its neuroinvasiveness and neurovirulence in mice are areas of concern , especially when considering the immunization of children and the immunocompromised . In this study , we present a novel MP-12-based , single-cycle replicable RVF vaccine candidate . This vaccine candidate was not neurovirulent in mice and was effective in protecting immunized mice from wild-type RVFV challenge , indicating its potential to be developed as a safe vaccine for use in both humans and livestock . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"viral",
"vaccines",
"viral",
"entry",
"rift",
"valley",
"fever",
"zoonoses",
"veterinary",
"diseases",
"viral",
"transmission",
"and",
"infection",
"zoonotic",
"diseases",
"virology",
"biology",
"microbiology",
"viral",
"diseases",
"viral",
"replication",
"veterinary",
"science"
] | 2014 | Development of a Novel, Single-Cycle Replicable Rift Valley Fever Vaccine |
The matrix ( MA ) domain of HIV-1 Gag plays key roles in membrane targeting of Gag , and envelope ( Env ) glycoprotein incorporation into virions . Although a trimeric MA structure has been available since 1996 , evidence for functional MA trimers has been elusive . The mechanism of HIV-1 Env recruitment into virions likewise remains unclear . Here , we identify a point mutation in MA that rescues the Env incorporation defects imposed by an extensive panel of MA and Env mutations . Mapping the mutations onto the putative MA trimer reveals that the incorporation-defective mutations cluster at the tips of the trimer , around the perimeter of a putative gap in the MA lattice into which the cytoplasmic tail of gp41 could insert . By contrast , the rescue mutation is located at the trimer interface , suggesting that it may confer rescue of Env incorporation via modification of MA trimer interactions , a hypothesis consistent with additional mutational analysis . These data strongly support the existence of MA trimers in the immature Gag lattice and demonstrate that rescue of Env incorporation defects is mediated by modified interactions at the MA trimer interface . The data support the hypothesis that mutations in MA that block Env incorporation do so by imposing a steric clash with the gp41 cytoplasmic tail , rather than by disrupting a specific MA-gp41 interaction . The importance of the trimer interface in rescuing Env incorporation suggests that the trimeric arrangement of MA may be a critical factor in permitting incorporation of Env into the Gag lattice .
Human immunodeficiency virus type 1 ( HIV-1 ) , like all replication-competent orthoretroviruses , encodes three main polyproteins – Gag , Pol and Env – which contain determinants necessary for particle assembly , enzymatic functions , and virus entry , respectively . HIV-1 assembly occurs in a series of steps , driven by the Gag precursor protein Pr55Gag ( for review , see [1] ) . HIV-1 Gag is comprised of four domains – matrix ( MA ) , capsid ( CA ) , nucleocapsid ( NC ) and p6 – and two spacer peptides located between CA and NC , and NC and p6 . Pr55Gag is able to form virus-like particles ( VLPs ) when expressed in cells in the absence of any other viral protein . The MA domain at the N-terminus of Pr55Gag directs cytoplasmic Gag to bind raft-like domains of the plasma membrane ( PM ) via specific recognition of phosphatidylinositol-4 , 5-bisphosphate [PI ( 4 , 5 ) P2] [2] ( for review , [3] ) . MA binding to PI ( 4 , 5 ) P2 , as well as Gag oligomerization , triggers a myristyl switch , exposing the myristic acid moiety covalently linked to the amino-terminus of MA [4] , [5] . The exposed myristic acid then inserts into the phospholipid bilayer , anchoring Gag to the PM . In addition to its PM-targeting function , MA is required for the incorporation of the viral Env glycoprotein complex into virions ( reviewed in [6] , [7] ) . Env is translated as a polyprotein precursor , gp160 , at the endoplasmic reticulum before it traffics to the PM through the Golgi apparatus , where it is cleaved into the mature surface glycoprotein gp120 and transmembrane glycoprotein gp41 . The mature Env glycoproteins remain associated as heterotrimers . Gp120 is located entirely on the exterior of the virion and mediates binding to the receptor ( CD4 ) and co-receptors ( CXCR4 or CCR5 ) , while gp41 anchors the Env complex in the lipid bilayer and mediates fusion between the viral and target cell membranes . HIV-1 gp41 , like the transmembrane glycoproteins of many lentiviruses , possesses a very long cytoplasmic tail ( CT ) . The long gp41 CT , which contains a variety of trafficking motifs ( for review see [8] ) , is required for the incorporation of Env into virus particles during assembly in physiologically relevant cell types such as CD4+ T cells and monocyte-derived macrophages ( MDMs ) , although it is not required for Env incorporation in some laboratory cell lines , such as HeLa [9] . Mutational studies support a direct role for the gp41 CT in Env incorporation and recent work has identified potential cellular trafficking proteins which influence Env incorporation in a CT-dependent manner [10] , [11] . Early in infection , the gp41 CT has also been reported to stimulate NF-kB , thereby enhancing virus replication in suboptimally activated target cells [12] . Reflecting its diverse roles in the virus replication cycle , mutations in MA can elicit a variety of defects . Single-amino acid mutations have been reported that block Env incorporation [13]–[15] . These mutations are noteworthy in that they typically do not impact any other aspect of the replication cycle and the infectivity block that they impose can be rescued , with the exception of 98EV , by Env glycoproteins bearing short CTs [13] , [15] , [16] . In 98EV short-tailed Env glycoproteins are incorporated but do not permit infectivity [15] . Mutation of the N-terminal Gly of MA , to which the myristate is covalently attached , or disruption of downstream residues that are either required for Gag myristylation or for the myristyl switch , impair Gag association with membrane [17]–[22] . Mutations in the highly basic patch of residues in the vicinity of residues 17–31 induce Gag retargeting to late endosomes as do mutations in the vicinity of residue 85 [19] , [23] . This mistargeting of Gag is thought to result primarily from a loss of MA-PI ( 4 , 5 ) P2 binding [2] , [5] ( for review see [24] ) . Mutations near residue 50 block virus assembly [19] . All or most of the MA domain can be deleted without blocking virus production if a membrane-targeting signal is attached to the N-terminus of Gag , but such mutants display promiscuous membrane targeting and impaired Env incorporation [25] . Finally , a small number of mutations have been shown to impair an early post-entry step in the replication cycle [26] , [27] . These mutations increase Gag-membrane association , suggesting that Gag affinity for membrane is fine-tuned to allow efficient virus assembly and , in the next round , virus entry and uncoating [26] , [27] . The mechanism of HIV-1 Env incorporation remains largely obscure . A variety of models have been proposed , including passive/random incorporation , co-trafficking of Gag and Env to a common site of assembly , direct interaction between Gag and Env , and an indirect interaction bridged by a cellular co-factor ( reviewed [6] , [7] ) . A central role for MA in Env incorporation is supported by a variety of findings , including the rescue of a gp41 CT mutant deficient in Env incorporation by a selected change in MA [10] and numerous examples of point mutations in MA that block Env incorporation but do not otherwise impair particle production [13]–[15] . As mentioned above , these MA mutants can typically be pseudotyped with alternative envelope glycoproteins that bear short CTs; e . g . , murine leukemia virus ( MLV ) Env , vesicular stomatitis virus ( VSV ) G glycoprotein , or HIV-1 Env mutants encoding a truncated gp41 CT [13] , [16] . These results confirm the original defect as one of Env incorporation and provide further evidence that there may be an interaction between the gp41 CT and MA – the domain of Gag proximal to the membrane and therefore best placed to interact with Env . However , neither the structure of the gp41 CT nor the structure of MA in the context of the virion has been established . The topology of the ∼150 amino acid gp41 CT with respect to the membrane has been the subject of controversy [28]–[30] ( for reviews see [8] , [31] ) ; it is thus currently unclear how much of the CT is exposed to MA on the cytosolic face of the membrane or what structure ( s ) the CT adopts . Relative to the gp41 CT , more is known about the structure of MA , as high-resolution structures have been generated by solution NMR and X-ray crystallography [32]–[34] . MA adopts a similar conformation in the NMR and crystal structures , but in the NMR structure MA is monomeric whereas MA forms a trimer in crystals [32] , [34]–[36] . Recent work has shown that MA organizes into hexamers of trimers on PI ( 4 , 5 ) P2-containing membranes in vitro [35] ( illustrated schematically in Fig . 1 ) . However , evidence for functional MA trimers in infected cells or virions is lacking , and cryo-electron tomography of mature and immature virions has likewise been unable to discern any long-range order within the layer of electron density corresponding to MA . There have been reports of direct interaction between Env and Gag in HIV-1 and simian immunodeficiency virus ( SIV ) systems; however , these results have proven difficult to reproduce and the nature of the putative Gag-Env interaction remains controversial [37]–[39] . In this study , we identify and characterize the mechanism of action of a MA substitution that is able to rescue a broad range of Env incorporation-defective mutants . Our data suggest that rescue depends on interactions between MA monomers at the trimer interface , providing evidence for the functional relevance of MA trimers in the immature Gag lattice . We propose that the trimeric arrangement of the MA domain of Pr55Gag plays an important role in HIV-1 Env glycoprotein incorporation into virus particles .
Previous studies demonstrated that a MA mutant , 34VE , is unable to incorporate Env into particles and consequently is unable to replicate efficiently in culture [14] , [20] . Prolonged culture of the 34VE mutant resulted in the acquisition of a second-site compensatory mutation in MA , 62QR , which reversed the Env incorporation , infectivity and replication defects of 34VE . In more recent analysis , we observed that passaging in the Jurkat T-cell line of another Env-incorporation-deficient MA mutant , 16EK [40] , again resulted in the acquisition of 62QR as a compensatory mutation ( Fig . 2A ) . To determine how broadly the 62QR mutation could rescue Env incorporation defects , we combined 62QR with a panel of mutations in MA and Env previously shown to block Env incorporation into particles . These included the MA mutations 12LE , 30LE , 98EV , and the gp41 d8 mutation , a five-amino-acid deletion in one of the helical domains of the gp41 CT [10] , [13] , [15] . The mutant molecular clones were transfected into Jurkat cells and virus replication was monitored . Each of the single MA mutants that had been previously identified as deficient for Env incorporation either failed to replicate or replicated with a significant delay relative to WT . By contrast , the MA double-mutant clones carrying 62QR all replicated with kinetics similar to those of the WT ( Fig . 2B ) . Perhaps most strikingly , 62QR was also able to rescue the replication defect induced by the d8 deletion in the gp41 CT ( Fig . 2B ) . To confirm that the loss of replication observed with the single mutants was due to loss of infectivity we performed single-cycle infectivity assays using the TZM-bl indicator cell line [41] . These experiments confirmed that the single-mutant viruses were unable to infect TZM-bl cells , whereas the double-mutant viruses carrying 62QR infected TZM-bl cells with an efficiency comparable to that of the WT ( Fig . 2D and F ) . Finally , to confirm that the lack of infectivity was due to loss of Env incorporation and that rescue by 62QR involved the restoration of Env incorporation , we collected viruses from transfected 293T cells and analyzed them by western blotting for the capsid ( CA ) protein and gp41 . Each of the single mutants displayed a lack of gp41 relative to WT , and in each case , double-mutant virions carrying 62QR contained WT levels of gp41 ( Fig . 2C and E ) . Collectively , these data demonstrate that all of the defective mutants tested can be rescued by 62QR , suggesting that their Env incorporation defects are caused by a similar mechanism . To address the mechanism by which 62QR rescues Env incorporation we examined Env incorporation into heterogeneous virus particles . A prevailing hypothesis for Env incorporation into HIV-1 particles posits a direct Gag-Env interaction [6] . If this were the mechanism of recruitment , then by making particles with a range of ratios of Env-recruiting ( e . g . , WT or 62QR ) and Env-excluding ( e . g . , 12LE ) Gags we should see Env incorporation vary in proportion to the amount of Env-recruiting Gag in the particle; no difference between WT and 62QR would be expected in this context ( a schematic representation of homogeneous and heterogeneous particles is shown in Fig . 3 ) . This experiment was performed in parallel with WT plus 12LE and 62QR plus 12LE molecular clones . As an increasing amount of the 12LE molecular clone was cotransfected with the WT clone , a rapid decline in virus infectivity was observed . By contrast , when 12LE and 62QR clones were cotransfected at the same ratios the effect was much less pronounced ( Fig . 4A ) . Indeed , even at a 3-fold excess of 12LE over 62QR , infectivity of 62QR:12LE virus was unaffected . It was not until 12LE was present at a 9-fold excess over 62QR that 62QR:12LE virus infectivity was comparable to that of virus produced at a 1∶1 WT∶12LE ratio . Similar analyses were performed , using a more limited range of input DNA ratios , with the mutants 16EK , 30LE , 34VE and 98EV with comparable results ( Fig . 4B–E ) . We performed an analogous set of experiments using the d8 gp41 mutant [10] . WT and 62QR molecular clones , both expressing the d8 Env mutant , were cotransfected over a range of DNA ratios . As shown in Fig . 4F , particles produced with WT Gag are poorly infectious; virions produced by 62QR Gag in the context of d8 Env are highly infectious . Even when WT DNA input was in 3-fold excess over 62QR , virus infectivity was not reduced by the d8 Env mutation ( Fig . 4F ) . It was not until WT was present in six-fold excess over 62QR that infectivity was reduced below 50% of that measured with 62QR alone ( Fig . 4F ) . These data demonstrate that the 62QR mutant , even when present at relatively low levels , is able to rescue , in trans , infectivity defects imposed by mutations in MA or the gp41 CT . Residue 62 lies in a region of MA that has not been previously implicated in Env incorporation; we therefore performed vertical scanning mutagenesis at this position to gain insight into its role in Env incorporation and the rescue of incorporation-defective mutants . We generated six additional mutants , 62Q[E/G/K/L/N/W] , and in parallel introduced each of these substitutions in the context of 12LE to look for rescue of the 12LE-imposed defect in Env incorporation . None of the single mutations at position 62 severely impaired virus release , infectivity or Env incorporation ( Fig . 5A , B and C ) . The most severe defect was seen in 62QW , which displayed approximately 50% reductions in Gag release and Env incorporation . All mutants replicated with WT kinetics in Jurkat cells ( Fig . 5D ) . The 12LE/62Q[E/G/K/L/N/W] double mutants were subjected to the same analyses described above for 12LE/62QR . Impaired virus release efficiency was observed for the double mutants 12LE/62QG , 12LE/62QN and 12LE/62QW . This defect in particle production is most likely due to an adverse effect on MA folding of introducing both 12LE and 62QG/N/W mutations , though we cannot exclude the possibility that these double mutants could be mistargeted . Unlike 62QR , none of the other residue 62 mutants was able to fully rescue the virus replication , infectivity , and Env incorporation defects imposed by 12LE ( Fig . 5 ) . The 12LE/62QK mutant exhibited a partial rescue , as infectivity in TZM-bl cells was comparable to that of 12LE/62QR ( Fig . 5C ) , and 12LE/62QK replicated sooner than the 12LE/62Q ( E/G/L/N/W] double mutants in Jurkat cells ( Fig . 5D ) . However , replication was delayed relative to that of the WT ( Fig . 5D ) and no rescue of Env incorporation was apparent ( Fig . 5B and C ) . Virus recovered from the 12LE/62QK cultures in these experiments had acquired 34VI or 34VL . 34VI is a previously characterized MA mutation that is capable of rescuing the Env incorporation and replication defects imposed by 12LE and d8 [10] , [14] . These data demonstrate that residue 62Q is not crucial for Env incorporation in the context of otherwise-WT MA , and that 62QR is unique among the mutants analyzed for its ability to fully rescue Env incorporation defects . The positions of the MA mutations that block Env incorporation were identified on the previously published MA crystal structures ( Fig . 6A and B; Fig . 3 ) [32] , [35] . The mutations in MA that affect incorporation of Env cluster towards the tips of the arms of the MA trimer; by contrast , the rescue mutation at position 62 is located near the center . As the gap in the MA lattice at the three-fold axis is relatively small and residue 62 is not prominently exposed to the membrane it seemed improbable that this site was engaging in direct contact with the gp41 CT ( Fig . 6A and B – in 6B the membrane would be at the top of the structure [5] ) . Instead , we hypothesized that the rescue depended on altering the structure of the MA lattice via novel interactions . The closest side chains to Q62 of chain a ( Fig . 6C ) are those of S66 and T69 in chain b; although these residues have polar side chains , the crystal structure indicates that they are too distant from Q62 to form hydrogen bonds , which typically require less than 3 Å between the nucleophilic atoms [42] ( Fig . 6C ) . It is possible , however , that when Q62 is replaced by R62 , a longer side chain combined with the greater positive charge may permit inter-subunit interactions to occur , perhaps with T69 ( Fig . 6D ) . A similar possibility exists with 62QK , although our data suggest it may be a less favored configuration ( Fig . 4; Fig . 6D and E ) . To test this hypothesis , the residues at positions 66 and 69 were mutated to Ala in the context of WT , 12LE , 62QR and 12LE/62QR . The 66SA mutation had no effect on the phenotypes of the four viral clones ( Fig . 7A ) . By contrast , 69TA blocked the ability of 62QR to rescue 12LE infectivity and Env incorporation , although 62QR/69TA was as infectious as 62QR alone ( Fig . 7A ) . 69TA as a single mutant was also impaired for infectivity and Env incorporation , suggesting that residue 69 may be involved in MA function in the WT molecule ( Fig . 7A ) . 69TA also showed a small ( 2-day ) delay in replication ( Fig . 8C ) , consistent with its reduced Env incorporation and single-cycle infectivity ( Fig . 7A ) . The partial defect of 69TA was relieved by 62QR ( Fig . 7A; Fig . 8C ) . To further examine the role of interactions in this region , an additional panel of mutants was generated by changing S66 and T69 to Arg . Strikingly , 66SR behaved like 62QR in its ability to rescue 12LE Env incorporation and infectivity , indicating that an Arg at either residue 62 or 66 could rescue the 12LE defect ( Fig . 7B; Fig . 6F ) . In contrast , combining 12LE , 62QR , and 66SR resulted in a loss of infectivity , suggesting either a loss of interaction or potentially repulsive interactions between R66 and R62 ( Fig . 7B; Fig . 6G ) . All virions bearing 69TR displayed very low infectivity and Env incorporation ( Fig . 7B ) ; structural modeling suggests that 69TR may introduce steric hindrance at the MA trimer interface ( Fig . 6H ) . These single-cycle results were confirmed by performing virus replication assays in Jurkat cells ( Fig . 8 ) . In each case , mutants that were found to be infectious were also able to replicate , although 12LE/66SR displays a moderate delay relative to WT and , like 12LE/62QK , 12LE/66SR acquires 34VI to permit efficient replication . Those mutants that were neither infectious nor able to replicate in Jurkat cells were pseudotyped with VSV-G ( Fig . 8F ) . The VSV-G pseudotyped particles were infectious in TZM-bl cells , confirming that the infectivity defect related to Env incorporation and not any other aspect of the infectivity process . As expected , these mutants could not be pseudotyped with HIV-1 Env ( Fig . 8F ) , consistent with the data presented above . Collectively , these data illustrate the importance of the MA-MA interface in the rescue of Env-incorporation defective mutants , suggesting that the multimeric arrangement of MA is a critical factor in HIV-1 Env incorporation .
Various models can be invoked to explain the incorporation of Env into HIV-1 particles , the principle unresolved issues being whether or not Env is actively recruited into virions and if so , whether Env interacts directly with Gag or indirectly via a bridging cellular factor [6] . Evidence to support active recruitment is provided by five observations . First , the existence of mutations in MA and Env that prevent Env incorporation is consistent with a direct recruitment and suggests the presence of interacting motifs , although it does not address the question of whether the interaction is direct or indirect . Secondly , it has been reported that HIV-1 Env is retained on immature particles even after removal of the viral membrane with detergent [43] . This retention is dependent on the long gp41 CT , again consistent with an interaction between the CT and Gag . Third , there have been reports of interaction between Gag and Env in cells and with recombinant proteins in vitro [37]–[39]; however , this interaction has been difficult to demonstrate consistently . Fourth , Env has been reported to influence the site of virus budding in polarized epithelial cells and T cells [44] , [45] . Finally , the observation that Gag processing ( virus maturation ) affects Env fusogenicity is consistent with cross-talk between Gag and the CT of gp41 [46]–[48] . In light of the difficulties encountered in reproducibly demonstrating a direct interaction between MA and Env , we sought an alternative approach to determine whether rescue by 62QR conformed to a model of Env recruitment via MA binding . We initially observed rescue in homogeneous particles in which all MA molecules contain both the defective and rescue mutations . If the defective mutants fail to incorporate Env because MA cannot bind the gp41 CT , then in particles composed of a mixture of Env-recruiting MA ( WT or 62QR ) and non-recruiting MA ( the defective mutants ) Env incorporation should vary in proportion to the amount of Env-recruiting MA the particle contains . In this scenario , WT and 62QR would function similarly in the heterogeneous particle assay; both should gradually become less infectious as a defective mutant is added to the particles . This was not observed . Rather , we observed a “dominant-positive” effect whereby particles containing 62QR retained infectivity and Env incorporation even when 62QR MA was a minority population compared to the defective mutant . WT-containing particles did not exhibit this phenotype , supporting the hypothesis that 62QR MA establishes a Gag structure that is more accommodating of the long gp41 CT . Although the structure of the gp41 CT is currently unknown , structures have been solved for MA [32] , [36] . The NMR structure shows an MA monomer , whereas the crystallographic structure reveals a trimeric arrangement . In the context of either structure the defective mutations are clustered , indicating the surface of MA that is important for Env packaging . If this site on MA actively binds Env then the rescue mutation might be expected to be nearby , and effect rescue by creating a new interaction to replace that lost by the defective mutant . Contrary to this hypothesis , 62QR is located at a site distant from the original mutations . Alternatively , 62QR could create an entirely new Env-binding site on the opposite side of MA . This hypothesis also faces multiple challenges . First , in the trimeric structure of MA , 62QR lies at the interface of the trimer , providing a much smaller gap in the MA lattice than that found in the center of the MA hexamer of trimers [35] ( Fig . 3 ) . Second , residue 62 is not present on the face of MA that would be expected to oppose the PM , so would require part of the gp41 CT to extend into the trimeric interface . In such a scenario it is hard to explain the lack of inhibitory phenotype of the non-conservative 62Q mutations reported here . Third , 62QR is able to rescue d8 , an Env mutant with a small deletion that is not efficiently packaged into WT particles . If this Env mutant had lost a MA-interacting motif then 62QR would be required to interact with an entirely new surface on Env . Our results do not exclude the possibility of a MA-Env interaction , but they suggest that it may not be required for Env recruitment into virions . The crystal structures of HIV-1 and SIV MA provide evidence for the existence of a MA trimer , and trimers of MA and Gag have been observed with recombinant proteins in vitro , although this trimer has not been observed directly in cells or virions [32] , [34] , [49] , [50] . The trimer hypothesis gained recent support from a lower-resolution approach that examined a more physiologically relevant two-dimensional array of MA on a phospholipid membrane . This analysis initially revealed MA hexamers , but when membranes containing PI ( 4 , 5 ) P2 were used , MA arranged itself as hexamers of trimers [35] , [51] ( Fig . 8 ) . In addition to supporting the trimer structure observed in the earlier crystals , these findings reveal a higher-order structure with potential relevance to Env packaging . The mutations that block Env incorporation cluster on the tips of the spokes of the MA trimer; in the context of the hexamer of trimers , this places them around the edge of a large hole in the proposed MA lattice ( Fig . 3 ) . It could be envisaged that the gp41 CT fits into this hole during particle assembly , and that the defective mutants are unable to package Env due to steric hindrance and/or charge repulsion resulting from mutations around the perimeter of this hole or mutations in gp41 CT , such as d8 , that may alter both its shape and orientation . This type of defect would explain the ability to packaged short tailed envelopes and could be rescued by a distant mutation if it were able to modify the structure of the MA lattice; 62QR is located near the trimer interface of the MA crystal structure and is therefore ideally placed to impose such a long-range change . The model that 62QR modifies the structure of the MA lattice could also explain the different phenotype of WT and 62QR in mixed Gag particles . Detailed examination of the crystal structure of the MA lattice revealed two potential partners for 62QR in forming inter-subunit interactions: S66 and T69 . These polar residues are too distant from Q62 to form hydrogen bonds , albeit only by 1–2 Å . The replacement of Gln with Arg in 62QR introduces a larger , charged side chain; if this interaction were sufficient to alter the structure of the MA lattice it could relieve the steric hindrance introduced by the defective mutations . Our data support this hypothesis , as mutation 69TA blocks the ability of 62QR to rescue 12LE , while other combinations of residues that could form interactions between the subunits , including 62Q with 66SR , are able to rescue Env incorporation and infectivity . Furthermore , introduction of multiple positively charged residues blocks rescue of 12LE , even where both mutations ( 62QR and 66SR ) are independently capable of rescue . This underscores the likelihood that the critical interaction is taking place between MA monomers and loss of rescue is due to the loss of intersubunit interactions . The apparent importance of interactions between MA monomers in the MA trimer over primary amino acid sequence raises important questions about the role of MA trimerization in HIV-1 Env incorporation . The putative first step in Env incorporation is the trafficking of both Gag and Env to lipid microdomains where assembly occurs [2] , [52]–[55] . These domains may be characterized by physical and biochemical features such as membrane curvature , distinct lipid and protein composition , and the presence of Gag itself [3] , [56]–[60] . These factors likely permit Env clustering at assembly sites without direct interaction with Gag , as foreign Env molecules have also been reported to cluster at budding sites [61] . Recent studies using super-resolution microscopy techniques demonstrate that Gag induces HIV-1 Env clustering and reduced Env mobility at sites of Gag assembly [62] , [63] . These effects of Gag assembly on Env clustering and mobility are largely dependent on the gp41 CT [62] , [63] . Mutations in MA that disrupt Env incorporation reverse the Gag-induced Env clustering [62] . Interestingly , Env clustering could be visualized in regions peripheral to Gag assembly sites , suggesting that clustering was not solely due to Gag-Env interaction but was also likely influenced by the formation of a Gag-induced microdomain that favored Env retention near the budding site [62] . Defects in Env incorporation induced by mutations in MA likely arise as a result of steric exclusion of Env from the assembled Gag lattice rather than a lack of recruitment to the budding site . Likewise , mutations in the gp41 CT ( e . g . , d8 ) probably alter the structure of the gp41 CT such that it is excluded from the Gag lattice . This steric exclusion model is supported by the observation that mutations in MA and gp41 such as those analyzed here block Env incorporation even in cell types such as HeLa that do not require the long gp41 CT for incorporation [9] , [10] . Also consistent with this hypothesis , large or charged side chains at position 12 in MA impair Env incorporation more dramatically than smaller , hydrophobic side chains [14] . If there is insufficient space in the Gag lattice to accommodate the gp41 CT , Env may be excluded and will diffuse away from the budding site . It is notable that the available structures of a 2D MA lattice are broadly similar , either hexameric in the absence of PI ( 4 , 5 ) P2 or a hexamer of trimers in the presence of PI ( 4 , 5 ) P2 [35] , [51] . The structures differ in the extent to which MA monomers pack together as trimers , effectively increasing the diameter of the gap in the center of the hexamer . It is possible that the role of MA trimerization is to create this larger gap in the MA lattice and thereby permit the long gp41 CT to pack into the lattice ( Fig . 9 ) . Such a hypothesis may explain the reduced Env incorporation exhibited by 69TR; modeling indicates that such a mutation would cause steric hindrance to the trimer structure , as it is currently understood . Mutations predicted to impair MA trimerization were previously found to reduce infectivity , although that study did not report loss of Env incorporation [64] . Further investigation of 69TR and other mutations that would likely destabilize the MA trimer will be invaluable in understanding the role this structure plays in particle assembly and Env incorporation . In summary , we have identified a MA mutation capable of global rescue of Env incorporation defects and have investigated the mechanism of rescue . The data reveal the importance of intersubunit interactions in a MA trimer , strongly supporting the existence of this structure in immature virions . Our data also suggest that the inability of several mutants to incorporate Env into particles may be caused by steric hindrance if the gap in the MA lattice is no longer large enough to accommodate the gp41 CT . If this is the case , then two possible drug targets can be proposed . Firstly , the sensitivity of Env incorporation to mutations at the tips of the MA trimer indicates that compounds binding at this site could inhibit Env incorporation . Secondly , the importance of intersubunit interactions within the MA trimer suggests that compounds able to disrupt intertrimer interactions may also inhibit Env incorporation , or potentially other functions of MA . Our data do not exclude the possibility of a direct interaction between MA and Env; they do , however , point to the importance of multimeric MA structure in Env incorporation . Determining the role of the MA trimer in HIV-1 particle assembly and replication remains a critical goal in extending both our understanding of basic retroviral biology and our range of virus-specific therapeutic targets .
HeLa and TZM-bl cells were cultured in Dulbecco's Modified Eagle's Medium ( DMEM ) , supplemented with 5% v/v fetal bovine serum ( FBS ) , 100 U/ml penicillin , 100 µg/ml streptomycin , and 2 mM L-glutamine ( Gibco ) . TZM-bl is a HeLa-derived indicator cell line that expresses luciferase following infection by HIV [65] . 293T cells were cultured in DMEM , supplemented with 10% v/v FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , and 2 mM L-glutamine . Jurkat CD4+ T-cells were cultured in Roswell Park Memorial Institute ( RPMI ) 1640 medium , supplemented with 10% v/v FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , and 2 mM L-glutamine . Anti-HIV-1 IgG is pooled patient serum obtained from the NIH AIDS Reagent Program . HIV-1 gp41 was detected with the 2F5 monoclonal antibody [66] . HIV-1 particles were generated using the full-length proviral clone pNL4-3 [67] . Point mutations were introduced by first subcloning the BssHI-SpeI fragment from pNL4-3 into pBluescript ( Stratagene ) . Mutations were introduced using the Quikchange method ( Stratagene ) following the manufacturer's instructions and the mutant BssHI-SpeI fragment was recloned into pNL4-3 . All mutations were confirmed by DNA sequencing ( Macrogen ) . For pseudotyping experiments VSV-G was expressed from pHCMVG , a gift from J . Burns ( University of California , San Diego ) . HIV-1 replication was assayed by rate of spreading infection in Jurkat cells as reported previously [19] . Virus replication was monitored by measuring RT activity as described [68] . When necessary , genomic DNA was extracted using QIAamp ( Qiagen ) following the manufacturer's protocol; proviral sequences were amplified by PCR and sequenced ( Macrogen ) . The efficiency of virus particle assembly and release was assayed by radiolabeling newly synthesized Gag then separately immunoprecipitating cell-associated and virion-associated Gag . Immunoprecipitated proteins were separated by SDS-PAGE and quantified using a Personal Molecular Imager ( Biorad ) [69] . The percentage of the total expressed Gag that is released from the cell as virion-associated material indicates the efficiency of particle release . For infectivity assays , virus-containing supernatants were generated by transfecting subconfluent 293T cells in 12-well plates . Shortly before transfection the medium was removed and replaced with 500 µl DMEM . 1 µg DNA was diluted in 500 µl serum-free DMEM , mixed thoroughly with 7 . 5 µl polyethyleneimine [1 mg/ml polyethyleneimine ( PEI ) , 20 mM HEPES pH 7 . 2] and incubated for 30 min at room temperature before dropwise addition to cells . Supernatants were harvested 24–48 h post-transfection and assayed for RT activity as described [68] . TZM-bl cells were infected with the supernatants and the luciferase signal was measured 24 h post-infection using Britelite Plus ( Perkin-Elmer ) . Infectivity was defined as the level of luciferase expressed by TZM-bl cells divided by the total amount of virus ( RT ) with which they were infected . Virions were harvested 24–48 h post-transfection by filtering supernatant through a 0 . 45 µm membrane then pelleting by centrifugation at 76 , 000×g for 1 h at 4°C . Virions were resuspended in 2× Laemmli buffer ( 120 mM Tris-Cl [pH 6 . 8] , 4% SDS , 20% glycerol , 10% β-mercaptoethanol , 0 . 02% bromophenol blue ) and analyzed by western blotting . Protein samples were separated by SDS-PAGE and transferred to a polyvinylidene fluoride ( PVDF ) membrane ( Immobilon , Millipore ) by semi-dry electroblotting . Membranes were probed with primary antibody overnight at 4°C , washed , then incubated for 1 h with species-specific horseradish peroxidase-conjugated secondary antibody . After the final washes , bands were revealed by chemiluminescence; working-solution was produced by mixing an equal volume of solution 1 [25 mM Luminol ( 3-aminophthalydazide ) , 0 . 3 mM p-coumeric acid , 100 mM Tris-HCl , pH 8 . 5] with solution 2 ( 0 . 01% hydrogen peroxide , 100 mM Tris-HCl , pH 8 . 5 ) . After incubation for 1 min at room temperature , membranes were exposed to a charge-coupled device in a Universal Hood II ( Biorad ) . For weak signals , Supersignal West Femto reagent ( Thermo Scientific ) was used following the manufacturer's instructions . Quantification was performed using ImageLab software ( Biorad ) . To calculate Env incorporation , volumes ( average pixel intensity multiplied by the area covered by the band ) were determined for each gp41 band and divided by the volume for the corresponding CA band . | One of the enduring problems in HIV-1 research is the mechanism of incorporation of the viral envelope ( Env ) glycoprotein into viral particles . Several models have been proposed ranging from an entirely passive process to a requirement for binding of Env by the matrix ( MA ) domain of the Gag precursor polyprotein . It is clear that specific regions within MA and Env play important roles , as mutations in these domains can prevent Env incorporation . We have identified a point mutation in MA that rescues a broad range of Env-incorporation defective mutations , located both in MA and in Env . Our investigations into the mechanism of rescue have revealed the importance of interactions between MA monomers at a trimeric interface . Our results are consistent with previously published crystallographic models and now provide functional support for the existence of MA trimers in the immature Gag lattice . Furthermore , as the modification of trimer interactions plays a role in the rescue of Env incorporation , we propose that MA trimerization and the organization of the MA lattice may be critical factors in Env incorporation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Global Rescue of Defects in HIV-1 Envelope Glycoprotein Incorporation: Implications for Matrix Structure |
Background: Calodium hepaticum ( syn . Capillaria hepatica ) is a worldwide helminth parasite of which several aspects of transmission still remain unclear . In the Amazon region , the mechanism of transmission based on the ingestion of eggs present in the liver of wild mammals has been suggested as the cause of the spurious infections described . We performed an epidemiological investigation to determine the incidence , risk of spurious infection and the dynamics of transmission of C . hepaticum in a community of the Brazilian Amazon . Methodology/Principal Findings: Stool samples of 135 individuals , two dog feces and liver tissue from a peccary ( captured and eaten by the residents ) were analyzed by conventional microscopy . Dog feces were collected from the gardens of households presenting human cases of spurious C . hepaticum infections . Community practices and feeding habits related to the transmission of the parasite were investigated . The individual incidence of spurious infection was 6 . 7% ( 95% CI: 2 . 08–11 . 24 ) . Cases of spurious infection were observed in 7 . 5% of the families and the household incidence was from 50% to 83 . 3% . The risk of spurious infection was 10-fold greater in persons consuming the liver of wild mammals ( p = 0 . 02 ) . The liver tissue of a peccary and one feces sample of a dog presented eggs of C . hepaticum . The consumption of the infected liver was the cause of the spurious infections reported in one household . Conclusions/Significance: This is the first identification of a source of spurious infection by C . hepaticum in humans and we describe a high rate of incidence in household clusters related to game liver alimentary habits . The finding of a dog feces contaminating peridomiciliary ground suggests the risk of new infections . We conclude that the mechanism of transmission based on the ingestion of liver is important for the dynamics of transmission of C . hepaticum in the studied area .
Calodium hepaticum ( syn . Capillaria hepatica ) is a zoonotic nematode of the Trichinellidae family found worldwide . This helminth infects the hepatic parenchyma of rodents ( principle hosts ) and various other mammals ( e . g . carnivores , humans ) of different families [1] . In humans infection may cause hepatic calodiasis ( syn . hepatic capillariasis ) , a rare liver disease ( 72 cases reported around the world , 5 being found in Brazil ) which may have a severe clinical course [1]–[5] . Infection by C . hepaticum occurs following the ingestion of embryonated eggs ( true or hepatic infection ) which pass through the intestinal tract . Larvae hatch at the level of the cecum , pass through the intestinal wall and reach the liver via the portal-hepatic system . The larvae mature in the hepatic parenchyma , transforming into adults 28 days after the infection . Females lay the eggs in the parenchyma and these develop only to the eight-cell stage . Eggs reach the environment through the decay of the host carcass or when a predator or cannibal ingests the host and releases the eggs through the stools . Over a 5–8 week period in optimal conditions of temperature , humidity and air exposure , the eggs embryonate in the ground and may infect a new host . Ingestion of non embryonated eggs leads to untrue ( or spurious ) infection in which the eggs pass through the intestinal tract and exit with the stools without causing liver disease [6]–[8] . The dynamics of the transmission of C . hepaticum and the risk factors associated with infection remain unclear [9] , [10] . In urban areas transmission is related to the presence of small rodents ( e . g . Rattus novergicus and Mus musculus ) and poor hygienic and sanitary conditions [1] , [6] , [11] . In small rodents , characteristics such as the high prevalence of natural infection [7] , [11] , [12] , the rapid populational turnover and the habit of cannibalism may explain the elevated transmission of the parasite among these rodents and their involvement in environmental contamination by eggs [13] , [14] . The ingestion of eggs present in the ground or in contaminated foods has been accredited as the mode of transmission to humans in urban areas . It has been suggested that domestic animals ( cats and dogs ) may also contaminate the peridomiciliary ground with infected stools [1] , [14] after eating small rodents , carcasses or infected liver of other mammals [15] . The participation of domestic animals in the domiciliary cycles has not , as yet , been well defined . Spurious infection has predominantly been described in tribal or immigrant communities around the world [5] . Several authors have suggested that the cause of this infection in determined populations is the mechanism of transmission based on the ingestion of non embryonated eggs present in the liver of mammals [15]–[20] . Foster & Johnson related the occurrence of spurious infection in natives of Panama to the encounter of three new hosts ( Tayassu pecari , Ateles geoffroyi and Cebus capucinus ) commonly used by the natives as food [16] . In a rural community in the Brazilian Amazon a case of spurious infection was associated with the reported consumption of liver of tapir [18] . Recently , 41 cases of spurious infection and the true infection of a peccary ( T . pecari ) and a monkey ( Ateles paniscus ) were reported in an indigenous amazonian population from Brazil suggesting the potential of these animals as local reservoirs [21] . However , studies are needed to confirm the mechanisms of transmission of C . hepaticum to humans as well as provide evidence of the cycles potentiating this transmission . More than half of the spurious infections by C . hepaticum reported worldwide in the last decade have been found in Brazil [5] . Ninety-eight percent ( 81/82 ) of these cases are from indigenous tribes or rural communities of the Amazon region ( from the States of Mato Grosso and Rondônia ) [17]–[19] , [21]–[25] . Nonetheless , no case of disease has , to date , been described in this region . The probable explanation is diagnostic difficulties in the Amazon that may be attributed to factors such as scarce access to health care services , unawareness of health professionals of the existence of the pathogen and the co-existence of tropical diseases ( such as malaria , viral hepatitis , arbovirosis , toxocariasis , among others ) [19] which share the same clinical symptoms and signs ( typical syndrome for C . hepaticum: persistent fever , hepatomegaly and leukocytosis with eosinophilia ) [14] suggesting that hepatic calodiasis is probably neglected in this region [18] , [19] . The aim of the present study was to determine the incidence and risk of spurious infection as well as the dynamics of transmission of C . hepaticum in a community in the Brazilian Amazon region .
This study was approved by the Ethics Committee in Investigation of the Oswaldo Cruz Foundation ( Protocol 384/07 of 20/08/2007 ) . Written Informed consent was obtained from all the study participants . According to the current regulations of the Brazilian legislation and of the Commission of Ethics in the Use of Animals ( CEUA ) of the Oswaldo Cruz Foundation , the study of dog feces samples collected from the gardens of households does not require ethical approval because the dogs were not handled or manipulated by the researchers . Dog owners provided prior permission for the collection of dog feces samples from their gardens . This study was carried out in the agricultural community of Rio Pardo of the municipality of Presidente Figueiredo , located ∼160 Km to the north of the city of Manaus ( ∼1°48′S; 60°19′W ) , Amazonas State , Brazil ( Figure 1 ) . This community was officially created in 1996 by the National Institute of Colonization and Agricultural Reform ( INCRA ) , in an area of tropical jungle . It is composed of 7 unpaved roads , known locally as “Ramal” , which includes households on both sides of these roads surrounded by tropical rain forest . The community also includes a riverine population living along the Rio Pardo stream known as “Igarapé” . A population census ( October–September of 2008 ) identified 701 inhabitants in the Rio Pardo community , with 360 ( 51 . 4% ) living in the Ramal area and 341 ( 48 . 6% ) in the Igarapé area . Most of the incomers are natives from the Amazon Region and make their livings from subsistence farming , plant harvesting ( wood , chestnuts , medicinal herbs ) , hunting and fishing . Most of the households present precarious basic sewage systems . Health care services are sparsely available in the community . A cross-sectional coproparasitologic study ( Text S1 ) of 40 randomly selected households was performed in the community in August 2009 . One stool sample was collected from each participant and evaluated 1–6 times by the Lutz [26] and/or Paratest ( Diagnostek , São Paulo , Brazil ) techniques . In addition , feces samples of dogs collected from the gardens of households presenting human cases of C . hepaticum and a liver tissue sample of a wild mammal ( captured and eaten by the residents ) were analyzed by the Lutz technique . The liver tissue was manually shredded in a NaCl solution at 0 . 85% prior to performing the diagnostic technique . Identification of the eggs of C . hepaticum was based on morphologic and morphometric analysis of 20–50 eggs per sample . The morphologic analysis was based on aspects of the structure of the eggshells [27] , [28] . Photomicrographs were made with a Leica microscope . A questionnaire was applied to obtain socio-demographic and epidemiologic information , especially community activities ( hunting ) , individual risk factors ( habits of ingestion of game meat ) and family practices related to the transmission of the parasite ( the habit of sharing game meat with dogs ) . The characteristics of the population and the eggs of C . hepaticum were described using tables of frequencies if the variables were qualitative and calculating means , standard deviations , maximum and minimum values if the values were quantitative . Comparisons of groups and the associations among variables were evaluated with Chi-square or Fisher exact tests . Estimations of incidence and relative risk ( RR ) were made with a 95% confidence interval ( CI ) . The analyses were performed using the SPSS v . 18 statistical package and the EPIDAT 3 . 1 . A level of significance of 5% was set .
Nine cases of spurious infection were identified , representing an incidence of 6 . 7% ( 95% CI: 2 . 08–11 . 24 ) . The eggs presented morphologic and morphometric characteristics compatible with the species of C . hepaticum , being yellowish-brown in color , barrel-shaped , with shallow polar plugs and radial striations and measuring an average of 64 . 4 µm in length and 36 . 7 µm in width ( Figure 2 , Table 1 ) . The cases were from households located in the area of Igarapé and in one of the Ramals of the community . Of the individuals infected , 55 . 5% were women and 55 . 5% children ( <14 years of age ) . The rate of households with spurious infection was 7 . 5% ( 95% CI: 1 . 50–20 . 38 ) . Eight out of nine ( 88 . 9% ) of the cases were found in two households of the Ramal . The rate of intradomiciliary spurious infection was 83 . 3% ( 5/6 ) in one household and 75% ( 3/4 ) in the other . All the cases were asymptomatic with the exception of two individuals in the same household who presented diarrhea and were both co-infected by Blastocystis hominis and Salmonella spp . The case of spurious infection from the area of Igarapé was an adult woman , the only participant that could not be found to do the questionnaire . In this latter case , the rate of intradomiciliary spurious infection was 50% . The habit of game intake was reported by 94 . 8% ( 127/134 ) of the individuals . The animals most frequently consumed were paca ( 85% ) , peccary ( 57 . 5% ) , armadillo ( 42 . 5% ) , agouti ( 37 . 5% ) and deer ( 37 . 5% ) . Game was eaten at least once a week by 25 . 6% , with the liver of game being eaten by 57 . 5% . The risk of spurious infection was 10-fold greater in those eating the liver of wild mammals [10% vs . 0% ( p = 0 . 02 ) ] . After undertaking the epidemiological investigation the complete history of the spurious infections in the Ramal was obtained . The residents reported that a few days prior to the coproparasitologic study a group of hunters captured several peccaries and shared the entrails and meat among the residents of the Ramal for food . The two families presenting cases of C . hepaticum reported having eaten the liver of the hunted peccaries . In addition , one of the families reported that raw meat remaining from the peccary liver that had been consumed was still stored in the freezer of their home . This piece of liver tissue was provided and analyzed in the laboratory , being positive for the presence of typical eggs of C . hepaticum . On average the eggs measured 63 . 1 µm in length and 36 . 3 µm in width ( Figure 2 , Table 1 ) . In this household the consumption of infected liver was the cause of the spurious infection reported in 83 . 3% ( 5/6 ) of the residents . The host was probably a Pecari tajacu or T . pecari since there are only two species of peccary in the study area . Some families reported the habit of giving game meat ( raw ) with their dogs as food . We estimated that 7 . 5% ( 3/40 ) of the families surveyed did this . Two dog feces samples were collected from the gardens of the two Ramal households presenting human cases of C . hepaticum . One of the samples analyzed presented eggs with characteristics compatible with species C . hepaticum , measuring an average of 61 . 1 µm in length and 35 . 4 µm in width . ( Figure 2 , Table 1 ) .
In the present study we describe a rate of spurious infection of 6 . 7% in a rural community of the Amazon , being , to our knowledge , one of the highest reported to date . This rate was similar to that estimated for indigenous people of the northwest of State of Mato Grosso ( 8 . 6% ) [21] and of the Suruí etnia in Rondônia ( 5 . 2% ) , in the Brazilian Amazon [25] , indicating that the Amazon region has the highest incidence of spurious infection worldwide . Other studies have reported lower rates ranging from 0 . 2% to 2 . 3% [17] , [22]–[24] . It should be noted that the rate estimated here might have been lower than that presented if all the samples had been evaluated only once . Three capillarid species of zoonotic importance are known: C . hepaticum , Eucoleus aerophilus ( syn . Capillaria aerophila ) and Paracapillaria ( Crossicapillaria ) philippinensis ( syn . Capillaria philippinensis ) [5] . E . aerophilus is widespread and parasitizes the trachea and mainly the bronchi of dogs , cats , wild carnivores and , occasionally , humans [29] , [30] . P . philippinensis is a parasite of fish , endemic in Philippines and Thailand and is the etiologic agent of human intestinal capillariasis [31] . Only the species C . hepaticum has been reported in Brazil . Eggs of C . hepaticum , E . aerophilus and P . philippinensis can be found in human feces and can be differentiated . In capillarids , different aspects of the eggshell structure can be used as a taxonomic clue [27] , [28] . The combination of morphologic and morphometric analysis of the eggs allows the identification of species of capillarids at a light microscopy level [27] , [32] . Morphologic characteristics of the bipolar plug ( asymmetric in E . aerophilus , inconspicuous flattened in P . philippinensis ) , the shell ( with a network of anastomosing ridges in E . aerophilus , striated in P . philippinensis ) and the shape ( peanut like in P . philippinensis ) can be used for differentiation of the eggs [30] , [33] . The morphology of the eggs found in this study ( from dog feces , human stools and liver tissue ) was compatible with the species C . hepaticum ( presence of shallow polar plugs and radial striations ) with dimensions according to those described by previous authors ( 40–75 µm in length×27–41 . 3 µm in width ) [8] , [14] , [18] , [21] , [27] , [34] . We report a frequent habit of wild mammal meat ( 94 . 8% ) and liver ( 57 . 5% ) intake similar to previous studies in Amazon populations and in indigenous tribes [18] , [19] , [25] . Recently , in a river-side population from the State of Rondônia ( western Brazilian Amazon ) with a high consumption ( 91 . 7% ) of meat from wild mammals ( paca , agouti or peccary ) , the serum prevalence of C . hepaticum was 34 . 1% at a dilution of 1∶150 , suggesting frequent contact with eggs of C . hepaticum [19] . Mild diarrhea has been reported in spurious infection of C . hepaticum , although this type of infection appeared to be asymptomatic in most cases [35] . In this study most individuals were asymptomatic , but the occurrence of diarrhea in two subjects could not be attributed to spurious infection by C . hepaticum due to the concomitant presence of two potential agents of diarrhea ( B . hominis and Salmonella spp . ) . This is the first report of a causative source of spurious infection of humans by C . hepaticum , that of peccary liver . Peccaries of the species T . pecari and P . tajacu are natural reservoirs of C . hepaticum [16] , [21] , [36] , are widely distributed in Brazil [37] , and are one of the wild mammals most frequently used as food in Brazilian amazonian communities [19] . For these reasons we suggest that these animals can be an important source of spurious infection for humans in the Amazon region . In Brazil , liver infection by C . hepaticum has been described in domestic dogs and cats and other mammals of the subfamilies Murinae ( R . novergicus , Rattus rattus and M . musculus ) , Sciurinae ( Sciurus aestuans ) , Caninae ( Lycalopex gymnocercus , Cerdocyon thous and Chrysocyon brachyurus ) , Tayassuinae ( P . tajacu and T . pecari ) , Felinae ( Puma concolor ) and Atelinae ( A . paniscus ) [21] , [34] , [36] , [38]–[40] . We estimated , for the first time , that individuals who usually eat the liver of wild mammals present a 10-fold higher risk of presenting spurious infection than those without this habit . As a consequence of this alimentary habit the spurious infection showed previously unreported high intradomiciliary rates ( 50% to 83 . 3% ) , characterized as infection by household clusters . The present results confirm the suspicion of several authors as to the existence of the mechanism of transmission by the ingestion of non embryonated eggs present in the liver of mammals and their involvement as a cause of spurious infection in humans . This thereby allows the conclusion that this is an important mechanism of transmission of eggs of C . hepaticum in this area and probably also in other areas of the Amazon with similar sociocultural characteristics . Eggs characteristic of the species C . hepaticum were found in a sample of dog feces collected from the garden of one household presenting cases of spurious infection . It is known that domestic dogs are susceptible to infection by C . hepaticum [40] , [41] and other capillarid species ( E . aerophilus and Eucoleus boehmi ) [32] . E . boehmi ( syn . Capillaria boehmi ) is a parasite of the nasal cavities and sinuses of wild canines ( e . g . foxes and wolves ) and domestic dogs , and its eggs can also be found in feces . Eggs present asymmetrical plugs , tiny pits on the surface of the wall and measure 50–60 µm×30–35 µm [32] . E . aerophilus have been described in dogs from Europe , North America and Australia and E . boehmi in dogs from Europe and North America [32] . Only the species C . hepaticum has been described in domestic dogs from Brazil . The spurious infection by C . hepaticum of a pet within a setting presenting human spurious infections has not been previously described . This finding may be related to the report of the families about having given raw game meat to the dogs . The practice of feeding pets with raw meat and close living relationships between humans and pets have previously been suggested as having an important role in the transmission of zoonotic pathogens [42] , [43] . This suggests that dogs may potentiate the emergence of a peridomestic cycle of C . hepaticum in this area . Since the dogs usually deposit their feces around the household , a new epizootic focus could be established very close to the family thereby increasing the risk of spurious and hepatic infections and even the development of cases of disease , especially among children . Children are more likely to be infected because of pica ( especially geophagia ) [5] . The deficient sanitary conditions in the community studied may be another important factor contributing to the risk of further infections . This last characteristic is common in rural communities which routinely hunt in the Amazon region [18] , [19] , [25] , suggesting the risk of the emergence of cases in other populations . We therefore recommend the implementation of an epidemiologic surveillance system for the diagnosis of spurious infection ( with correct microscopic identification of the parasite ) in areas in which the population has the habit of eating game meat . To prevent mix-ups , laboratory technicians could be trained to differentiate the eggs of Trichuris trichiura from those of capillarids [5] , taking into account morphologic and morphometric characteristics . Since Trichuris spp . eggs have smooth walls they can be distinguished from the mainly ornamented eggs of the capillarids [44] . Moreover , in areas presenting spurious infections , we recommend the investigation of C . hepaticum in subjects with clinical suspicion of hepatic disease by serology and , if necessary , histopathological examination of liver biopsy samples [5] . To date , there are no molecular tools for the detection of C . hepaticum . As measures of prevention it should be recommended that families should cook the liver well prior to ingestion and should not feed dogs with raw entrails . Improvements in local sanitary conditions should also be implemented . Investigation of the sources of infection in areas in which the presence of spurious infection has been confirmed is advisable , including the mammals most frequently consumed and small rodents . In the latter case , several studies have described the adaptation of some small rodents ( Rhipidomys spp . and Mesomys spp . ) to villages and households located in deforested areas of the Amazon invaded by man [45] , [46] . Thus , their role in the dynamics of peridomiciliary transmission in rural Amazon areas should also be evaluated . In addition , the species M . musculus and R . rattus , which are widely distributed reservoirs of C . hepaticum in Brazil ( that adopts the human household or its proximities as its habitat ) , have already been described in an area of the Amazon biome with recent human occupation [47] . Near the location of the present study , in an area of forest reserves ( Minimum Critical Size of Ecosystems reserves ) , small rodents of some subfamilies , such as Sigmodontinae ( e . g . Euryoryzomys macconnelli , Hylaeamys megacephalus and Rhipidomys nitela ) and Eumysopinae ( e . g . Proechimys cuvieri ) have been found . Moreover , known C . hepaticum reservoirs , such as peccaries ( P . tajacu and T . pecari ) , A . paniscus and P . concolor have been described in the area [48] . This is the first study to identify a source of spurious infection of C . hepaticum in humans ( peccary liver ) in a rural community of the Brazilian Amazon . A high rate of incidence in household clusters is described in relation to the habit of the ingestion of liver of wild mammals . The finding of contaminated peridomiciliary ground with an infected dog feces suggests greater risk of new infections without the participation of a wild agent . The dynamics of transmission found in the community studied led to the conclusion that the mechanism of transmission following the ingestion of liver of wild mammals is an important mechanism in this area . | The zoonotic parasite Calodium hepaticum is the causative agent of rarely reported liver disease ( hepatic calodiasis ) and spurious infections in humans . In spurious infections eggs of this parasite are excreted in the stools without causing disease . It has been suggested that the cause of this type of infection in Amazonian areas is the ingestion of liver of wild mammals infected with the eggs of the parasite . Nonetheless , studies are needed to confirm this mechanism of transmission and investigate its epidemiological importance . In the present study we report the high individual ( 6 . 7% ) and household incidence ( 50%–83 . 3% ) of spurious infection in a rural community of the Brazilian Amazon . We found a high risk of spurious infection among subjects who usually ate the liver of wild mammals and detected a source of spurious infection in humans ( peccary liver ) as well as , for the first time , ground contamination with infected dog feces in a household presenting human cases . We confirm the existence of this mechanism of transmission of C . hepaticum and suggest that it is important for transmission not only in this area but probably also in other areas of the Amazon with similar sociocultural characteristics . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"soil-transmitted",
"helminths",
"epidemiology",
"environmental",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"neglected",
"tropical",
"diseases",
"parasitic",
"diseases",
"helminth",
"infection"
] | 2012 | Calodium hepaticum: Household Clustering Transmission and the Finding of a Source of Human Spurious Infection in a Community of the Amazon Region |
Placental malaria ( PM ) can lead to poor neonatal outcomes , including low birthweight due to fetal growth restriction ( FGR ) , especially when associated with local inflammation ( intervillositis or IV ) . The pathogenesis of PM-associated FGR is largely unknown , but in idiopathic FGR , impaired transplacental amino acid transport , especially through the system A group of amino acid transporters , has been implicated . We hypothesized that PM-associated FGR could result from impairment of transplacental amino acid transport triggered by IV . In a cohort of Malawian women and their infants , the expression and activity of system A ( measured by Na+-dependent 14C-MeAIB uptake ) were reduced in PM , especially when associated with IV , compared to uninfected placentas . In an in vitro model of PM with IV , placental cells exposed to monocyte/infected erythrocytes conditioned medium showed decreased system A activity . Amino acid concentrations analyzed by reversed phase ultra performance liquid chromatography in paired maternal and cord plasmas revealed specific alterations of amino acid transport by PM , especially with IV . Overall , our data suggest that the fetoplacental unit responds to PM by altering its placental amino acid transport to maintain adequate fetal growth . However , IV more profoundly compromises placental amino acid transport function , leading to FGR . Our study offers the first pathogenetic explanation for FGR in PM .
Pregnant women living in malaria endemic regions are highly susceptible to malaria , especially in first pregnancies [1] , [2] . Malaria in pregnancy is characterized by placental malaria ( PM ) , the selective accumulation of Plasmodium-falciparum infected erythrocytes ( IE ) in the maternal intervillous blood space of the placenta , in direct contact with the nutrient-transporting epithelium , the syncytiotrophoblast . When placental malarial infection is poorly controlled , chemokine release results in the recruitment of maternal immune cells , predominantly monocytes , to the intervillous blood spaces [3] . The resultant inflammation is termed intervillositis ( IV ) [4] . In comparison to PM without local inflammation , PM with IV is associated with significant decreases in birthweight and an increased prevalence of low birthweight ( LBW ) deliveries , primarily due to fetal growth restriction ( FGR ) [1] , [2] , [5] , [6] . Recent studies have begun to shed light on the pathogenetic mechanisms linking PM and FGR ( reviewed in [7] ) . Inadequate maternal nutrition and placental insufficiency have been proposed . In Congolese women studied by serial ultrasound examinations , FGR associated with PM was 2–8 times more common in undernourished than in well-nourished mothers [8] . The same undernourished mothers with PM had increased uterine artery resistance ( Griffin et al . submitted ) , which is associated with placental insufficiency . A decreased fetal/placental weight ratio is one manifestation of placental insufficiency found in primigravid women with PM [9] . It has previously been suggested [9] , [10] that FGR and LBW associated with PM could be caused by impaired capacity of the placenta to transport maternal nutrients , especially amino acids , to the growing fetus . Although this postulate has never been formally tested , it is supported by observations in idiopathic FGR showing that the activities of various placental nutrient transporters are selectively altered [11] , [12] . Among the nutrient transporters affected is system A , a group of Na+-dependent neutral amino acid transporters that actively transfer small , neutral amino acids and thereby enables the establishment of high intracellular amino acid concentrations , which are then used to exchange for extracellular essential amino acids via system L [13] , [14] . In the placenta , system A activity is mediated by three Na+-dependent neutral amino acid transporter ( SNAT ) isoforms belonging to the SLC38 gene family; SNAT1 ( SLC38A1 ) , SNAT2 ( SLC38A2 ) and SNAT4 ( SLC38A4 ) . All isoforms are expressed on the microvillous plasma membrane ( MVM ) of the human syncytiotrophoblast [15] . A reduced system A amino acid transporter activity in MVM has been consistently observed in placentas of pregnancies associated with FGR [16]–[18] , and the reduction in system A activity in MVM correlates well with the severity of FGR [19] . Further , various animal studies have suggested that reduced system A activity may be causally related to the etiology of FGR [20]–[22] . Pro-inflammatory cytokines produced by monocytes have been shown to decrease system A activity . IL-1β reduces system A activity in trophoblast cells [23] and acute exposure to TNF-α resulted in diminished maternofetal transfer of a system A analogue in a rat model of FGR [24] . These cytokines have been associated with LBW in PM , especially when associated with IV [25]–[27] , and their production could be caused by PM , either through activation of monocytes by IE [28] or by direct effects of IE on syncytiotrophoblast leading to secretion of cytokines and chemokines [29] , [30] . This suggests a link between PM , IV and altered placental amino acid transport , with impacts on fetal growth and development . In the current study , we hypothesized that the release of soluble mediators triggered by IV associated with PM impairs placental transport of amino acids across the syncytiotrophoblast , contributing to the pathogenesis of FGR and LBW . We found that PM , especially with IV , was associated with decreased placental amino acid uptake and dysregulated maternofetal amino acid balance , likely to alter the transfer of amino acids to the fetus , and to contribute to the pathogenesis of PM-associated FGR .
Characteristics of the individuals who participated in the various aspects of the study are summarized in Table 1 . SLC38A1 transcript levels were reduced ( p = 0 . 008 ) in the syncytiotrophoblast of infected placentas with IV compared to that of uninfected placentas , while levels in syncytiotrophoblast of infected placentas without IV were intermediate . A similar trend was observed for SLC38A2 transcript levels ( Fig . 1A ) . SLC38A1 ( p = 0 . 017 ) but not SLC38A2 ( p = 0 . 39 ) transcript levels were lower in the syncytiotrophoblast of placentas of LBW infants compared to normal birthweight infants ( Fig . 1B ) . Figure 2A reveals that Na+-dependent MeAIB uptake by MVM vesicles from infected placentas either with or without IV was lower ( p≤0 . 015 ) than uptake by vesicles from uninfected placentas . Na+-dependent MeAIB uptake was similar between groups with PM ( p = 0 . 65 ) . Birthweight was positively associated with Na+-dependent MeAIB uptake by MVM vesicles from all placentas ( Rho = 0 . 26 , p = 0 . 07; Fig . 2B ) . In response to P . falciparum infection , monocytes elicit a pro-inflammatory response including the secretion of IL-1β [28] . IL-1β has been previously reported to decrease Na+-dependent MeAIB uptake by placental trophoblast cells [23] . We therefore investigated whether PM with IV was associated with increased IL-1β concentration and if conditioned medium from a monocyte/IE co-culture could impair Na+-dependent MeAIB uptake . IL-1β plasma concentration in maternal blood harvested from placentas with PM and IV was higher ( p = 0 . 017 ) compared to uninfected controls , and comparable ( p = 0 . 1 ) to the PM without IV group ( Fig . 3A ) . Within the group of PM with IV , IL-1β concentration was negatively correlated with birthweight ( Rho = −0 . 52; p = 0 . 04; Fig . 3B ) . Because IL-1β is produced by monocytes in response to IE [28] and because Na+-dependent MeAIB uptake by MVM vesicles was lowest in PM with IV ( Fig . 2A ) , we speculated that system A activity impairment could be attributable to products generated by monocytes in response to IE . Medium collected from a monocyte/IE co-culture was used to mimic the intervillous space milieu in cases of PM with IV . This medium was applied to human placental choriocarcinoma BeWo cells to investigate its effect on Na+-dependent MeAIB uptake . Cell viability was monitored in all subsequent experiments and was unaffected by any of the treatments ( data not shown ) . An inhibition of Na+-dependent MeAIB uptake by BeWo cells when exposed to monocyte/IE co-culture conditioned media was consistently observed ( Fig . 4A ) . An IL-1β blocking antibody was used to investigate the role of IL-1β in mediating this effect ( Fig . 4B ) . At the concentration used , the blocking antibody was effective in abolishing recombinant IL-1β-mediated reduction in Na+-dependent MeAIB uptake . In contrast , addition of IL-1β blocking antibody had no effect on the inhibition of Na+-dependent MeAIB uptake observed with monocyte/IE conditioned media . This indicated that IL-1β was not a major factor in the reduction in Na+-dependent MeAIB uptake observed with monocyte/IE conditioned media . Treatment of BeWo cells with uninfected erythrocytes ( UE ) or with lysed or intact IE that had or had not been opsonized by human Ig did not alter Na+-dependent MeAIB uptake by BeWo cells compared to media control ( p≥0 . 12; data not shown ) . This suggests that the reduction in Na+-dependent MeAIB uptake by BeWo cells relies on monocytes' response to IE more than effects of IE per se . We next investigated the potential effect of altered system A activity , indicated by the reduced Na+-dependent MeAIB uptake observed in PM with IV , on fetal amino acid levels by measuring free amino acid concentration in paired maternal and cord plasma samples ( Table 2 ) . A low fetal/placental weight ratio is a marker of placental insufficiency and has been associated with malaria [9] , [31] , [32] and PM with IV cases had lower fetal/placental weight ratio than uninfected controls ( p = 0 . 036 ) . For a number of amino acids , cord concentration was positively correlated with fetal/placental weight ratio , either among all PM cases or for those with IV ( Table 3 ) . Among babies with PM and IV , cord concentrations of several neutral , branched chain amino acids , transported by system L [13] were positively associated with fetal/placental weight ratio . This suggests that intervillositis may lead to placental insufficiency in part through impaired transplacental amino acid transport . There was no positive correlation between cord concentration of amino acids and birthweight ( p≥0 . 29 ) .
Understanding the pathogenesis of PM-associated FGR is critical to the design of novel interventions to decrease its burden [33] . In this study , we identified an impaired placental amino acid uptake and dysregulated maternofetal amino acid balance in PM , especially with IV , providing a pathogenic mechanism for PM-associated FGR through altered transfer of amino acids to the fetus . The activities of various transport mechanisms in the plasma membranes of the syncytiotrophoblast are dysregulated in idiopathic FGR [34] , [35] . In particular , the activity of system A amino acid transporters has often been reported to be downregulated in FGR [17]–[19] , and animal studies demonstrate that a reduction in system A activity precedes development of FGR [22] . We observed reduced transcription of the system A transporters SLC38A1 ( SNAT1 ) , and , to a lesser extent SLC38A2 ( SNAT2 ) , within the syncytiotrophoblast in PM with IV , which is compatible with the reduction we observed in system A activity , and consistent with the involvement of these two SNAT subtypes in the downregulation of system A activity associated with PM with IV . Previous studies of idiopathic FGR have not demonstrated altered SLC38A1 or SLC38A2 transcription in whole placental lysates [36] , whereas here we have specifically measured SLC38A1 and SLC38A2 transcript levels from the syncytiotrophoblast . Our data suggest that in cases of PM with IV , the syncytiotrophoblast responds to infection and inflammation by down-regulating the transcription of these SNAT isoforms . As activity of SNATs is partly regulated at the level of transcription [37] , this suggests that PM with IV may decrease SNAT-mediated placental amino acid transport . To investigate this further , we studied system A activity ex vivo . System A activity was decreased in PM alone , and to a greater extent in PM with IV . The relationships observed between PM or birthweight with system A activity and SNAT transcript levels suggest that system A makes important contributions to fetal growth , and that these contributions are compromised by PM , especially PM with IV . To understand how PM , especially with IV , might impair placental amino acid transport we developed an in vitro model to examine trophoblast cell responses to factors present in the placental intervillous blood space . Regardless of the way they were presented to placental cells , IE alone did not induce a significant decrease in system A-mediated MeAIB uptake . This is in accord with our ex vivo data , and with clinical observations that PM without IV is not associated with FGR [6] . In contrast , conditioned media from monocyte-IE co-cultures , which mimic the intervillous milieu in PM with IV , significantly reduced MeAIB uptake by trophoblast cells , indicating that monocytes participated in eliciting this response . Our evidence suggests that IE activate monocytes to release factors that inhibit system A activity . A number of factors have been shown to modulate system A activity in placental cells or BeWo layers including cytokines such as IL-1β , IL-6 and TNF-α [23] , [38] which are increased in PM [25]–[27] and produced by monocytes in response to IE [28] . Despite IL-1β concentrations being raised in maternal blood of the intervillous space in PM with IV and negatively correlating with birthweight in this group , the decreased Na+-dependent MeAIB uptake by BeWo cells in our in vitro model was not substantially mediated by IL-1β , as illustrated by the inability of blocking antibody to IL-1 β to counteract the inhibitory effect of monocyte-IE co-culture supernatants; the mediator ( s ) responsible are at present unknown and could either be a factor ( s ) consumed by malaria-stimulated monocytes or a factor ( s ) secreted by these cells . The cause of the decreased amino acid uptake observed ex vivo in MVM from women with PM and IV is not known . As discussed above , it may be mediated by monocyte-derived factors , but these remain to be conclusively identified . Hormones that stimulate system A-mediated amino acid uptake including IGFs [39] and leptin [40] are decreased in PM [41] , [42] , and these may contribute in part to the decreased system A activity demonstrated in patient samples . In vitro , supernatants from co-cultures ( rather than monocytes themselves ) inhibit amino acid uptake , suggesting that local depletion of available amino acids by activated monocytes is not a significant contributing factor . We next assessed whether the observed decrease in system A activity in PM , or effects of PM on other amino acid transport systems , resulted in altered amino acid concentrations in maternal and cord blood . In normal pregnancy , delivery of some amino acids , particularly essential amino acids , is only just sufficient to meet fetal requirements [43] , [44] . In pregnancies compromised by severe FGR , maternofetal transfer of amino acids may be reduced [45] , [46] . In PM without IV , maternal concentration of a number of amino acids ( Asn , Asp , Cys and Trp ) was increased compared to uninfected controls , possibly due to reduced uptake of these amino acids by the syncytiotrophoblast , resulting in increased maternal concentrations . In PM with IV , maternal concentration of all amino acids except Ala was either unchanged or elevated compared to uninfected controls , consistent with observations in idiopathic FGR [46] . Cord Ala levels were also lower compared to uninfected controls and positively correlated with maternal levels in PM with IV ( Rho = 0 . 56; p = 0 . 002 ) . Thus , malaria-related inadequate maternal concentrations of amino acids were not responsible for changes in fetal amino acid concentrations . We did not see widespread decreases in cord amino acid concentration in the group with PM and IV , as have been described in idiopathic FGR [19] , [45] . This lack of widespread impact of PM on cord amino acid concentration could be explained by the degree of severity of the FGR in our study compared to the idiopathic FGR studies . In the latter , birthweight was dramatically decreased , by ∼600 g to ∼1550 g . In contrast , in our cohort , birthweight of control infants was relatively low , and birthweight only differed by ∼200 g between control infants and those with PM and IV ( in keeping with larger epidemiological studies in this population [5] , [27] ) . In resource-poor settings such as Malawi , obstetric care is limited , and women with at risk pregnancies due to highly compromised placental function and severe FGR may not be identified for intensive management , but may instead experience pregnancy loss . Malaria is a common cause of stillbirth and miscarriage in such settings [47] , [48] , and our study design may have resulted in malaria-affected pregnancies with severe FGR being under-represented in our cohorts . Longitudinal studies of at-risk pregnancies may be useful in quantifying the risks and manifestations of severe FGR in malaria-affected pregnancies further . Differences in cord blood amino acid concentrations between groups suggest that placental transport and/or metabolism of a number of amino acids is altered in PM . In the PM without IV group , the neutral and anionic classes of amino acids were most notably affected , suggesting a selective effect on placental handling of these amino acids . In infected women without IV but not in the group with PM with IV , there was a particularly striking increase in the fetal concentration of the anionic amino acids Asp and Glu , which are taken up into the placenta from the maternal and fetal circulations respectively by system XAG− . The physiological significance of this is unclear at present and it could suggest that IV restored placental transport of these amino acids . However , it is known that Glu uptake from the fetal circulation plays a crucial role in fetoplacental Glu-Gln cycling and may also serve to protect the fetus against Glu neurotoxicity [49] . The similar trend observed for Asn , Gly , Met ( system A substrates ) and both Met and Leu ( system L substrates ) also implicates these systems as being affected differentially in PM without IV as against PM with IV . Whether this reflects altered amino acid transport and/or placental amino acid utilization or production has yet to be established . The decrease in system A activity could also indirectly impair the activity of other systems that depend on the gradient of amino acid concentration established by system A for their own activity . Trp was the only amino acid for which there was a failure to concentrate in cord blood as compared to maternal , in the PM with IV group . In idiopathic FGR , Trp concentration occurs [45] implying that the failure of Trp to concentrate in our cases was related to the presence of PM rather than FGR per se . Other essential amino acids ( Ile , Leu , Phe , Thr , Val ) , which , like Trp , are transported predominantly by system L [50] , were concentrated in cord blood as compared to maternal , suggesting that there was no global impairment of system L activity . In placental infections , Trp is catabolised through the kynurenine pathway , notably by the enzyme indoleamine 2 , 3 dioxygenase [51] , and we speculate that similarly increased placental catabolism of Trp in PM contributes to the lack of Trp placental concentrative capacity we observed . Cord Trp concentration was increased in PM without IV , suggesting that Trp placental catabolism may be reduced with acute infection; a change that is then blocked by inflammatory cells in chronic infection . Taken together , our data suggest that PM alters placental function through effects on multiple amino acid transporter systems , and that these effects are selective for certain amino acids; PM may also increase placental amino acid metabolism . In PM with IV , the dysregulation of maternofetal amino acid concentrations is more pronounced , possibly because the monocytes accumulating in the intervillous space release inflammatory mediators that alter the activity of amino acid transporters in the syncytiotrophoblast . We have shown effects of PM on one amino acid transport system , system A , in vitro and ex vivo , and found clues for the dysregulation of other placental amino acid transport systems . Amino acid transport is highly complex , with overlapping and interdependent pathways . Although the defects we observed in amino acid transporter activity did not translate directly into lower fetal amino acid concentrations in women with PM and IV , we did observe important correlations in women with PM , or PM and IV , between low fetal/placental weight ratio , an index of placental insufficiency , and low cord levels of critical amino acids . Our evidence calls for studies to further characterize the effects of PM and IV on the activity of system A as well as investigating the activities of systems XAG− and L in the placenta . Such studies should also capture whether PM with IV alters transplacental transport of glucose [52] or lipids which , together with amino acids , form essential substrates for fetal growth [53] , [54] . In order to counteract the decrease in placental nutrient transport , nutrient supplementation interventions [55] could be implemented , but should ideally be combined with further research to ensure that such interventions correct , and do not exacerbate [56] , defects in transport and fetal growth [57] . Greater understanding of the mechanisms by which PM affects placental nutrient transport , combined with possible interventions to improve fetal growth in malaria , are important priorities in areas of the world where the co-existence of malaria and maternal malnutrition threaten the health and lives of millions of young babies .
From 2001–2006 , pregnant women delivering a live singleton newborn in the labor ward of Queen Elizabeth Central Hospital , Blantyre , Malawi were recruited into a case-control study . Cases were defined by the presence of P . falciparum asexual parasites on placental blood smear . For each case identified , two uninfected , age ( ±2 years ) and gravidity-matched controls , negative for malaria parasites by both peripheral and placental smears , were then enrolled . Inclusion and exclusion criteria have been described elsewhere [41] . The College of Medicine Research Ethics Committee , University of Malawi , approved the study and written informed consent was obtained from all participants . Immediately after delivery maternal and cord venous blood were collected and separated by centrifugation . Plasma was stored at −80°C . One set of placental biopsies was snap-frozen in liquid nitrogen for MVM purification , another set was embedded in optimal cutting temperature ( OCT ) medium before being frozen at −80°C for laser capture microdissection ( LCM ) of the syncytiotrophoblast and a last set was fixed in 10% neutral-buffered formalin for malaria infection grading . Placental tissue sections were examined by light microscopy for presence of malaria infection . In infected placentas , 500 randomly-selected intervillous space maternal blood cells were counted as previously described [5] , to derive estimates of placental parasite density and monocyte counts ( expressed as percentage of all maternal intervillous cells ) . Samples used in the study were selected from the cohort of participants based on tissue availability , after assessment of placental histology as described below . Presence of IE in the intervillous space of the placenta defined PM cases . These were sub-grouped into PM with or without IV . IV was defined as a monocyte count ≥5% of all intervillous cells counted [5] . Uninfected placentas were defined as showing no signs of malaria infection or intervillositis . RNA was extracted from laser-captured syncytiotrophoblast as previously described [58] . Briefly , tissue cryosections immobilized on SuperFrost PLUS slides ( Fisher ) were air-dried and fixed in acetone . After rehydration , sections were stained with methyl green ( Sigma-Aldrich ) and dehydrated . Material captured by laser microdissection using a MicroBeam microscope ( P . A . L . M . Microlaser Technologies ) was catapulted directly into RNA extraction buffer ( RLT buffer with β-mercaptoethanol; Qiagen ) , and RNA extracted using an RNeasy Micro Kit ( Qiagen ) , according to the supplier's recommendations . Purified RNA was eluted and kept at −80°C . RNA ( 10 ng ) was reverse transcribed using Superscript III enzyme mix ( Invitrogen ) with random hexamers . Transcript levels for SLC38A1 ( SNAT1 ) , SLC38A2 ( SNAT2 ) and tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein , zeta polypeptide ( YWHAZ ) were quantified using 1∶4 dilution of cDNA for all samples . Primer sequences are shown in Table 4 . Real-time quantitative PCR was performed using previously reported thermal cycling conditions at an annealing temperature of 60°C with SYBR Green 1 ( Applied Biosystems ) [59] . Transcript levels were quantified against a standard curve generated from a pool of all placental cDNA samples . Preliminary studies confirmed YWHAZ transcript levels were comparable between groups and YWHAZ was used to normalize target gene transcript levels . Isolation of MVM vesicles from placental biopsies ( 7 . 2±1 . 6 g ) was performed using magnesium precipitation and differential centrifugation based on the method of Glazier et al . [60] as described previously [61] with modifications according to Jimenez et al . [62] to allow simultaneous recovery of the basal plasma membrane for other studies . Frozen placental biopsies were thawed and homogenized in 250 mM sucrose , 10 mM HEPES-Tris , pH 6 . 95 ( Buffer D; 2 . 5 volumes biopsy weight ) and a sample of the homogenate ( 1 ml ) was retained for further analysis . The remaining homogenate was centrifuged at 10000 g for 15 min at 4°C and the supernatant retained . The pellet was resuspended in buffer D ( 1 . 5 volumes of initial biopsy weight ) and the centrifugation step was repeated . The supernatants were pooled and centrifuged at 125000 g for 30 min at 4°C . The pellet was resuspended in buffer D and 12 mM MgCl2 added and stirred on ice for 20 min . The suspension was centrifuged at 2500 g for 10 min at 4°C . The supernatant ( containing MVM ) was centrifuged at 125000 g for 30 min at 4°C . The pellet was resuspended in 300 mM sucrose , 20 mM Tris-maleate , pH 7 . 4 and loaded onto a discontinuous 25%–37%–45% sucrose gradient . After centrifugation at 90000 g for 6 h at 4°C , the MVM fraction at the 37%–45% interface was recovered and centrifuged at 110000 g for 30 min at 4°C . The resultant MVM pellet was resuspended in intravesicular buffer ( 290 mM sucrose , 5 mM HEPES , 5 mM Tris-HCl , pH 7 . 4; 3 volumes pellet weight ) and repeatedly passed through a 25-gauge needle to vesiculate the MVM fragments to form vesicles . MVM vesicles were stored at −80°C . Protein concentration of placental homogenate and MVM vesicles was determined by the Lowry method [63] . Purity of MVM vesicle preparations was assessed by enrichment of alkaline phosphatase activity as described previously [60] . Alkaline phosphatase enrichment factors ( mean ± SD ) were not different ( p = 0 . 39 ) between uninfected ( 16 . 8±8 . 6; n = 18 ) , PM ( 11 . 5±6 . 0; n = 14 ) and PM with IV ( 12 . 9±6 . 7; n = 21 ) groups , suggesting comparable MVM purity between groups . 14C-MeAIB ( NEC-671; PerkinElmer , ) was used as a well-characterized , non-metabolizable amino acid analogue substrate to measure the activity of system A amino acid transporter [15] , [17] . Placental blood was aspirated from an incision made in the basal plate at a pericentral site of the placenta . Plasma was separated and frozen at −80°C . IL-1β was measured by ELISA ( DuoSet , R&D ) in undiluted plasma samples according to the manufacturer's instructions . Samples from primigravidae were preferentially selected for the amino acid analysis , as they are at highest risk of malaria in pregnancy [2] . In paired maternal and cord blood plasma samples , free forms of the common 20 amino acids were analyzed by reversed phase ultra performance liquid chromatography ( RP-UPLC ) with pre-column derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate ( AccQ Tag Ultra; Waters Corporation ) . Standards ( Amino Acid Standard H ( Pierce ) together with Gln , Trp and Asn ( Sigma ) ) were prepared with norvaline ( Sigma ) included as an internal standard . For assay , 100 µl plasma was mixed with an equal volume of 200 µM norvaline and deproteinated by ultrafiltration through a 10 kDa MWCO spin filter ( Millipore ) at 4800 g for 60 min at 10°C . 10 µl filtrate was derivatized and analyzed on a Waters Acquity UPLC over a 20 min gradient using a 2 . 1×150 mm , 1 . 7 µm i . d . , BEH C18 column ( Waters Corporation ) flowing at 0 . 6 ml/min at 60°C . Detection was via UV ( 260 nm ) and data was collected and analyzed using the Waters Empower2 software . Non-normally distributed data are presented as box plots showing the median , 25th/75th and 10th/90th centiles unless described otherwise . Values out of the 10th centiles were included in statistical analyses but not represented in graphs . Normally distributed values are presented as mean and standard deviation . Non-normally distributed data were normalized by log-transformation prior to statistical analysis . Data were then compared between 3 groups using one-way ANOVA . When the p value of the ANOVA test was lower than 0 . 1 , two-group comparisons were made using a 2-tailed T-test . Correlations were assessed using Pearson's correlation test . Trends across ordered groups were tested using Cuzick's test . The ability of the placenta to concentrate amino acids in cord blood was examined by testing if the cord to maternal concentration ratio of each amino acid was different from 1 using a one-sample T-test . The College of Medicine Research Ethics Committee , University of Malawi , approved the study and written informed consent was obtained from all participants prior to inclusion in the study . | Malaria infection during pregnancy can cause fetal growth restriction and low birthweight associated with high infant mortality and morbidity rates . The pathogenesis of fetal growth restriction in placental malaria is largely unknown , but in other pathological pregnancies , impaired transplacental amino acid transport has been implicated . In a cohort of Malawian women and their infants , we found that placental malaria , especially when associated with local inflammation , was associated with decreased expression and activity of an important group of amino acid placental transporters . Using an in vitro model of placental malaria with local inflammation , we discovered that maternal monocyte products could impair the activity of amino acid transporters on placental cells . Amino acid concentrations in paired maternal and cord plasmas revealed specific alterations of amino acid transport by placental malaria , especially with local inflammation . Overall , our data suggest that , more than malaria infection per se , the local inflammation it triggers compromises placental amino acid transport function , leading to fetal growth restriction . Greater understanding of the mechanisms involved , combined with interventions to improve fetal growth in malaria , are important priorities in areas of the world where the co-existence of malaria and maternal malnutrition threatens the health and lives of millions of young babies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"obstetrics",
"and",
"gynecology",
"pregnancy",
"obstetrics",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"parasitic",
"diseases",
"malaria",
"pathogenesis"
] | 2013 | Plasmodium falciparum Malaria Elicits Inflammatory Responses that Dysregulate Placental Amino Acid Transport |
Mitogen-activated protein kinase ( MAPK ) and PUF ( for Pumilio and FBF [fem-3 binding factor] ) RNA-binding proteins control many cellular processes critical for animal development and tissue homeostasis . In the present work , we report that PUF proteins act directly on MAPK/ERK-encoding mRNAs to downregulate their expression in both the Caenorhabditis elegans germline and human embryonic stem cells . In C . elegans , FBF/PUF binds regulatory elements in the mpk-1 3′ untranslated region ( 3′ UTR ) and coprecipitates with mpk-1 mRNA; moreover , mpk-1 expression increases dramatically in FBF mutants . In human embryonic stem cells , PUM2/PUF binds 3′UTR elements in both Erk2 and p38α mRNAs , and PUM2 represses reporter constructs carrying either Erk2 or p38α 3′ UTRs . Therefore , the PUF control of MAPK expression is conserved . Its biological function was explored in nematodes , where FBF promotes the self-renewal of germline stem cells , and MPK-1 promotes oocyte maturation and germ cell apoptosis . We found that FBF acts redundantly with LIP-1 , the C . elegans homolog of MAPK phosphatase ( MKP ) , to restrict MAPK activity and prevent apoptosis . In mammals , activated MAPK can promote apoptosis of cancer cells and restrict stem cell self-renewal , and MKP is upregulated in cancer cells . We propose that the dual negative regulation of MAPK by both PUF repression and MKP inhibition may be a conserved mechanism that influences both stem cell maintenance and tumor progression .
Mitogen-activated protein ( MAP ) kinases ( MAPKs ) control many aspects of animal development , including cell proliferation , differentiation , and survival [1] . Most relevant to this work are MPK-1 , the primary Caenorhabditis elegans MAPK/ERK homolog [2 , 3] , as well as ERK2 and p38α , two human MAPK homologs [1] . MAPK enzymes are activated by a class of dual specificity kinases that phosphorylate both threonine and tyrosine residues ( e . g . , [4] ) and are inactivated by a class of dual specificity phosphatases , called MAPK phosphatases ( MKPs ) ( e . g . , [5 , 6] ) . Aberrant ERK2 activation contributes to human developmental disorders , such as Noonan syndrome , Costello syndrome , and cardiofaciocutaneous syndrome ( reviewed in [7] ) ; p38α , on the other hand , is thought to inhibit tumor initiation by inducing apoptosis in response to oxidative stress [8] . In mouse embryonic stem cells ( mESCs ) , ERK2 and p38α MAPK signaling promotes differentiation and inhibits self-renewal [9 , 10] . The C . elegans germline provides a superb model for understanding the molecular controls of stem cells , proliferation , differentiation , and survival [11] . In this simple tissue , germline stem cells are restricted to the distal “mitotic region . ” At a molecular level , germline stem cells are maintained by Notch signaling and two RNA-binding proteins , fem-3 binding factor ( FBF ) -1 and FBF-2 . FBF-1 and FBF-2 ( collectively called FBF ) are nearly identical and largely redundant proteins that belong to the broadly conserved family of PUF RNA-binding proteins [12 , 13] . PUF proteins inhibit gene expression by binding regulatory elements in the 3′ untranslated region ( 3′UTR ) of their target mRNAs , thereby controlling their translation or stability [14] . FBF maintains germline stem cells by repressing mRNAs that encode differentiation-promoting regulators . For example , FBF represses gld-1 and fog-1 mRNAs , which encode regulators that promote entry into meiosis or sperm differentiation , respectively [15 , 16] . The role of PUF proteins in stem cell maintenance appears to be a conserved and perhaps ancestral function [14 , 17] , but the target mRNAs responsible for this function have not yet been identified . Here , we suggest that MAPK mRNA is one key target . Once C . elegans germ cells have left the mitotic region , they move proximally and progress through meiosis and gametogenesis . Activated MAPK controls exit from meiotic pachytene and physiological apoptosis during oogenesis [18 , 19] . Normally , about half of the germ cells progress from pachytene into diakinesis and develop as oocytes , and the other half of the germ cells undergo apoptosis in the proximal pachytene region [19]; however , in mutants with blocked MAPK signaling , germ cells arrest in pachytene and fail to die [18 , 19] . An antibody that specifically detects activated MAPK , called α-DP-MAPK ( for dually phosphorylated MAPK ) , reveals a dramatic increase in activated MPK-1 just prior to the pachytene to diplotene/diakinesis transition [20 , 21] . A key inhibitor of C . elegans MPK-1 is LIP-1 , a homolog of the dual specificity phosphatase MKP [22] . LIP-1 has two roles in germline development . First , LIP-1 controls the extent of germline proliferation in the mitotic region: wild-type germlines contain significantly more mitotically dividing germ cells than do lip-1 null mutants [23] . Because the depletion of mpk-1 rescued the lip-1 proliferation defect , it seems likely that LIP-1 promotes germline proliferation by inhibiting MPK-1 activity . Second , LIP-1 promotes the G2/M arrest typical of diakinesis , apparently by inhibiting MPK-1 activity after germ cells have exited pachytene [21] . In this paper , we explore the regulatory relationship between PUF proteins and MAPK expression , both in the C . elegans germline and in human embryonic stem cells ( hESCs ) . We find that mpk-1 mRNA is a direct target of FBF repression in C . elegans and that two human MAPK mRNAs , those encoding ERK2 and p38α , are repressed by human PUM2 . We also demonstrate that FBF and LIP-1 function redundantly to inhibit germ cell apoptosis and suggest that this dual regulation of MAPK signaling , which occurs at post-transcriptional and post-translational levels , respectively , may be conserved during diverse cellular processes in animal development and tissue homeostasis .
The mpk-1 gene encodes two major transcripts , mpk-1a and mpk-1b , which produce MPK-1A and MPK-1B proteins , respectively [2 , 24] ( Figure 1A ) . To identify which products were expressed in the germline , we performed RT-PCR of RNA prepared from adults that either contained a normal germline ( GL+ ) or contained no germline ( GL− ) . The mpk-1a mRNA is contained entirely within mpk-1b , but mpk-1b harbors a unique exon ( Figure 1A ) . We therefore examined mpk-1 mRNAs using either mpk-1ab primers , which recognize both isoforms , or mpk-1b−specific primers . The mpk-1ab mRNA was abundant in both GL+ and GL− animals , but mpk-1b mRNA was very low or undetectable in GL− animals ( Figure 1B ) . Therefore , mpk-1b appears to be enriched in the germline . To corroborate this result , we examined the two MPK-1 proteins in Western blots of protein prepared from wild-type ( GL+ ) , GL− mutants , and mpk-1 ( ga117 ) mutants . MPK-1A was present in both GL+ and GL− animals , but MPK-1B protein was found only in GL+ animals ( Figure 1C ) . We conclude that mpk-1b RNA and its MPK-1B protein are predominantly expressed in the germline . We next investigated the distribution of mpk-1 mRNA and MPK-1 protein in the germline . After in situ mRNA hybridization of extruded germlines , both mpk-1ab and mpk-1b−specific probes were low in the mitotic region , increased in the transition zone , and became abundant in the pachytene and oogenic regions ( Figure 1D and 1E ) . No signal was detected with the control mpk-1 sense probe ( Figure 1F ) . For immunohistochemistry , we used an anti-MAPK/ERK polyclonal antibody that cross-reacts with both MPK-1 isoforms in wild-type animals but is absent from mpk-1 ( ga117 ) mutants ( Figure 1C ) . The distribution of MPK-1 protein was similar to that of mpk-1 mRNA: MPK-1 protein was low in the distal germline ( e . g . , mitotic region , transition zone ) , was increased in the proximal pachytene region , and became abundant in developing oocytes ( Figure 1G and 1H ) . Essentially no signal was seen in mpk-1 ( ga117 ) mutant germlines ( Figure 1I and 1J ) . To investigate the isoform expressed , we depleted mpk-1b mRNA by RNA interference ( RNAi ) ; the specific elimination of MPK-1B was verified by Western blots ( unpublished data ) . MPK-1 protein was essentially absent from mpk-1b RNAi germlines , except for a low signal in developing oocytes ( Figure 1K and 1L ) . We conclude that MPK-1B is the predominant MPK-1 isoform in the germline . To determine if FBF might repress mpk-1 expression , we compared the abundance of MPK-1 protein in germlines that either had wild-type FBF ( both fbf-1 and fbf-2 ) or no FBF ( neither fbf-1 nor fbf-2 ) . For this study , we could not examine a simple fbf-1 fbf-2 double mutant , because that animal does not maintain mitotically dividing germ cells [15] . Instead , we examined mpk-1 expression in tumorous ( Tum ) germlines that have robust germ cell proliferation both with and without FBF . In gld-1 mutants ( Tum+FBF ) , MPK-1B was about 8-fold lower than in gld-1; fbf-1 fbf-2 mutants ( Tum−FBF ) ( Figure 2A , lanes 2 and 3 ) . MPK-1B was also lower in gld-1 gld-2 mutants ( Tum+FBF ) than in gld-1 gld-2; fbf-1 fbf-2 mutants ( Tum−FBF ) ( Figure 2A ) . By contrast , MPK-1A levels were equivalent in these strains ( Figure 2A ) . Therefore , FBF affects MPK-1B , but not MPK-1A , abundance . To visualize where within the germline FBF affects MPK-1 expression , we stained dissected germlines with both the MPK-1 polyclonal antibody and DAPI , and we quantitated levels with ImageJ software . Consistent with the Western blot data , MPK-1 was lower in gld-1 germlines than in gld-1; fbf-1 fbf-2 germlines ( Figure 2B–2F ) . This difference was particularly striking within the mitotic region , where MPK-1 was about 5-fold lower in Tum+FBF than in Tum−FBF germlines ( Figure 2B–2F ) . We also stained fbf-1 single mutant germlines , which maintain a mitotic region but are compromised for full FBF activity; in about 20% of dissected germlines , MPK-1 protein was detected in the distal mitotic region ( unpublished data ) . We conclude that FBF maintains a low level of MPK-1 protein in the distal germline . FBF binds specifically to FBFbinding elements ( FBEs ) within the 3′UTR of its direct target mRNAs [12 , 13 , 15 , 16 , 23 , 25]; potential FBEs can be recognized by a consensus sequence ( UGURHHAUW ) [“R , ” purine; “H , ” not G; “W , ” A or U] [26] . The mpk-1 3′UTR possesses two potential FBEs that conform to this sequence ( Figure 3A ) . To assess FBF binding to these predicted mpk-1 FBEs , we used both yeast three-hybrid ( Figure 3B and 3D ) and gel retardation assays ( Figure 3E ) . Yeast three-hybrid interactions were monitored by production of β-galactosidase from a lacZ reporter ( Figure 3D ) . The mpk-1 FBEa and FBEb interacted with both FBF-1 and FBF-2 in three-hybrid assays ( Figure 3C and 3D ) and bound to purified recombinant FBF-2 in gel shift assays ( Figure 3E ) . Furthermore , those interactions were specific: wild-type mpk-1 FBEa and FBEb bound FBF , but not PUF-8 ( Figure 3D ) or PUF-5 ( unpublished data ) , and that binding was disrupted by mutations of the UGU in the consensus binding site ( Figure 3C–3E , FBE* mutant changed UGU to aca ) . The apparent Kd values for mpk-1 FBEa and FBEb were about 93 nM and 320 nM , respectively . We conclude that the mpk-1 3′UTR bears two FBEs and that FBEa appears to have higher affinity for FBF than does FBEb . We next asked whether FBF protein associates with mpk-1 mRNA in the nematode . Specifically , we prepared C . elegans extracts from animals carrying either a rescuing epitope-tagged GFP::FBF or a control GFP::tubulin ( TUB ) , and incubated those extracts with immobilized GFP antibodies to immunoprecipitate ( IP ) associated mRNAs . We then used RT-PCR to assess either mpk-1 or control mRNAs ( eft-3 , negative control; gld-1 , positive control ) . mpk-1 mRNA was reproducibly enriched in the IP from GFP::FBF-bearing animals compared to that from the GFP::TUB animals ( Figure 3F ) . Therefore , FBF is likely to bind directly to the mpk-1 mRNA in vivo . Interestingly , the mpk-1 FBEa is conserved in three Caenorhabditis species: C . elegans , C . briggsae , and C . remanei ( Figure 3G ) . We conclude that the mpk-1 3′UTR possesses FBEs and that FBF repression of mpk-1 expression is direct . The C . elegans homolog of MAPK phosphatase , LIP-1 , behaves genetically as an inhibitor of MAPK activity and is likely to inactivate MPK-1 in germ cells ( see Introduction ) [21 , 23] . Therefore , MAPK is negatively regulated in the germline by two distinct mechanisms: FBF represses mpk-1 expression ( present work ) and LIP-1 inhibits MAPK activity . To test the possibility that FBF and LIP-1 might function redundantly to control the distribution of activated MPK-1 , we used the α-DP-MAPK monoclonal antibody , which recognizes the active form of MAPK by its dual phosphorylation ( DP ) . In wild-type germlines , activated MPK-1 was not detected in the distal germline ( e . g . , mitotic region , transition zone ) but became abundant in the proximal part of the pachytene region and in maturing oocytes ( Figure 4A ) [20 , 21] . A similar distribution was seen in fbf-1 and lip-1 single mutants ( Figure 4A–4C ) . By contrast , activated MPK-1 was broadly distributed in fbf-1; lip-1 double mutant germlines , extending all the way to the distal tip ( Figure 4D ) . We conclude that activated MPK-1 is subject to two redundant modes of downregulation: FBF acts post-transcriptionally to repress mpk-1 mRNA and LIP-1 is likely to act post-translationally to inhibit MPK-1 activity . In wild-type C . elegans hermaphrodites , physiological germ cell apoptosis requires MPK-1 activation [19] . Strong loss-of-function mutations in any of the genes of the RAS/MPK-1 pathway interrupt germ cell apoptosis [19] , but germ cell apoptosis does not increase in let-60/Ras gain-of-function ( gf ) mutants [19 , 27] . Although MPK-1 activity is much higher in let-60 ( gf ) germlines than in wild-type , the distribution of activated MPK-1 is similar in let-60 ( gf ) and wild-type germlines [21] . We hypothesized that germ cell apoptosis might be regulated by the distribution of activated MPK-1 rather than its quantity at the site of apoptosis . To test this idea , we counted the number of germ cell deaths in adult hermaphrodite germlines , using the vital dye SYTO 12 ( Molecular Probes ) to detect apoptotic corpses . In these experiments , we used fbf-1 mutants to deplete but not eliminate FBF activity: FBF-2 provides sufficient FBF to maintain germline stem cells . Wild-type hermaphrodite germlines had about 2 . 6 germ cell corpses per gonad arm ( Figure 4E , 4F , and 4K ) , and a similar number was seen in fbf-1 and lip-1 single mutants ( Figure 4K ) . However , in fbf-1; lip-1 double mutants , the number of germ cell corpses increased dramatically ( Figure 4G , 4H , and 4K ) . To determine if the increased germ cell apoptosis in fbf-1; lip-1 mutants depends on MPK-1 activity , we used mpk-1 ( ga111 ) , a temperature-sensitive mutation . Specifically , we compared the number of germ cell corpses after shifting fbf-1; lip-1 double mutants and fbf-1; lip-1; mpk-1 ( ga111ts ) triple mutants to restrictive temperature ( 25 °C ) . Whereas the fbf-1; lip-1 double mutant displayed excess germ cell death , the fbf-1; lip-1; mpk-1 ( ga111ts ) triple mutant had far fewer corpses ( Figure 4I , 4J , and 4K ) . Therefore , MPK-1 activity is required for the increased apoptosis in fbf-1; lip-1 double mutants . We conclude that FBF and LIP-1 proteins act redundantly to inhibit MPK-1 activity and promote germ cell survival . We next investigated the possibility that PUF RNA-binding proteins might also control MAPK expression in humans . This idea was inspired in part by the knowledge that the human PUF protein , PUM2 , is expressed abundantly in both human embryonic stem cells ( hESCs ) and human germline stem cells [28] and in part by the conserved PUF role in stem cell maintenance ( see Introduction ) . Predicted Pumilio binding elements ( known as NREs [nanos response elements] ) were sought using UGUANAU as a core consensus [29] . The Erk2 3′UTR possesses a putative NRE immediately adjacent to the cleavage and polyadenylation hexanucleotide sequence ( AAUAAA ) ( Figure 5A and 5B ) , and the p38α 3′UTR has four putative NREs ( Figure 5A ) . In yeast three-hybrid assays , the Erk2 NRE , p38α NREa , and p38α NREb all interacted specifically with PUM2 ( Figure 5C ) . Moreover , the wild-type NREs in Erk2 and p38α bound purified PUM2 protein in gel shift assays ( Figure 5D ) , but mutant NREs ( NRE* ) with an altered consensus ( Figure 5B ) did not ( Figure 5D ) . We next asked whether an Erk2 NRE is conserved in mouse Erk2 3′UTR . Intriguingly , the Erk2 NRE is conserved in human and mouse 3′UTRs , both being located next to the hexanucleotide sequence ( Figure 5E ) . We conclude that PUM2 protein binds to Erk2 and p38α 3′UTRs and that PUF binding to MAPK 3′UTRs is highly conserved . To test if PUM2 controls Erk2 and p38α expression , we performed a series of enhanced green fluorescent protein ( EGFP ) -based reporter assays in hESCs . Specifically , we fused an EGFP reporter to the Erk2 3′UTR that contained either a wild-type NRE , Erk2 3′UTR ( wt ) , or a mutated NRE , Erk2 3′UTR ( mut ) ( Figure 6A ) . We transfected these constructs along with a transfection control into hESCs and monitored EGFP expression 24 h later . We first observed EGFP using fluorescence microscopy and then determined expression levels by Western blot analysis ( Figure 6B–6J ) . As a control , we used a reporter carrying a 3′UTR without any predicted NREs ( EGFP::BGH [bovine growth hormone] 3′UTR ) . hESCs carrying the EGFP::BGH 3′UTR reporter expressed EGFP at a very high level ( Figure 6J ) . By contrast , hESCs transfected with the Erk2 3′UTR ( wt ) reporter had much less EGFP ( Figure 6B , 6C , and 6J ) . To ask if the NRE is critical for this low expression , we assayed Erk2 3′UTR ( mut ) , a reporter with three altered nucleotides in the NRE consensus ( UGU to aca ) ( Figure 6A ) . This Erk2 3′UTR ( mut ) reporter produced about 9-fold more EGFP than the Erk2 3′UTR ( wt ) reporter ( Figure 6D , 6E , and 6J ) . We speculated that endogenous PUM2 might repress expression of the Erk2 3′UTR ( wt ) reporter but not the Erk2 3′UTR ( mut ) reporter . Attempts to use siRNA to silence endogenous PUM2 were not successful . We therefore cotransfected hESCs with the EGFP reporters and PUM2::FLAG ( Figure 6A ) , and we found that PUM2::FLAG dramatically repressed Erk2 3′UTR ( wt ) expression ( Figure 6F , 6G , and 6J ) but did not repress Erk2 3′UTR ( mut ) expression ( Figure 6H , 6I , and 6J ) . We next asked if reporters carrying the p38α 3′UTR were also controlled in an NRE-dependent manner . To this end , we transfected hESCs with either of two EGFP-based reporter genes , p38α 3′UTR ( wt ) or p38α 3′UTR ( mut ) ( Figure 6A ) . As found for the Erk2 reporter , the wild-type , but not the mutant , p38α 3′UTR was capable of efficiently repressing expression from the EGFP reporter gene in hESCs ( Figure 6K–6O ) . In this case , expression from p38α 3′UTR ( wt ) was about 6-fold lower than that from p38α 3′UTR ( mut ) ( Figure 6O ) . Taken together , we conclude that the PUM2 binding elements present in both Erk2 and p38α 3′UTRs mediate repression in hESCs .
Both PUF RNA-binding proteins and MAPK enzymes are highly conserved from yeast to humans . In this paper , we show that PUF proteins directly bind to 3′UTR regulatory elements in MAPK-encoding mRNAs and thereby control the generation of MAPK protein . Specifically , C . elegans FBF binds and regulates mpk-1 expression in germ cells , and human PUM2 binds and regulates Erk2 and p38α 3′UTRs in hESCs . Similarly in yeast , the Mpt5 PUF protein inhibits Ste7/MAPKK expression to regulate the filamentation-specific MAPK pathway [30] . Therefore , an ancient relationship appears to exist between the PUF RNA-binding proteins and MAPK signaling . To our knowledge , our work provides the only direct link between PUF proteins and MAPK-encoding mRNAs . Because this direct connection exists in both C . elegans and humans , we suggest that it may represent a broadly conserved regulatory relationship among metazoans . Figure 7A places PUF repression into a conserved pathway of MAPK control . Specifically , PUF proteins are responsible for the post-transcriptional repression of MAPK mRNAs; mechanistically , this could be achieved by controlling either their translation or stability . PUF proteins were originally thought to control mRNA stability in yeast but to control mRNA translation in animals [14] , but as more examples of PUF-controlled mRNAs have surfaced , it has become clear that this generalization is too simple . For example , C . elegans FBF controls the stability of lip-1 mRNA [23] , and yeast Mpt5 controls the translation of HO mRNA [31] . Regardless of mechanism , our work shows conclusively that PUF proteins are direct regulators of MAPK-encoding mRNAs . MAPK is a key regulator of programmed cell death , among its other roles during animal development [1] . In this work , we investigated the function of PUF repression and MKP inhibition in the control of apoptosis in the C . elegans oogenic germline . In wild-type animals , about half of the germ cells die and the other half begin oocyte maturation ( Figure 7B ) [19] . Indeed , activated MPK-1 is most abundant where germ cells either die or begin oogenesis [20 , 21] . In mutants lacking either FBF-1/PUF or LIP-1/MKP , the distribution of activated MPK-1 is essentially normal and the number of germ cells that undergo cell death is also normal . By contrast , in double mutants lacking both FBF-1 and LIP-1 , activated MPK-1 extends all the way to the distal end of the germline , where it is normally never seen , and apoptosis increases dramatically . This result suggests two things . First , because distribution of activated MAPK affects number of apoptotic germ cells , the decision to die may be programmed at a location distal to their actual site of death . Second , and perhaps most important for this work , the distribution of activated MAPK and the number of germ cell deaths are controlled redundantly by FBF repression and LIP-1 inhibition . MAPK inhibition is ensured in the C . elegans germline , at least in part because FBF represses lip-1 mRNA in addition to its control of mpk-1 mRNA ( Figure 7B ) [23] . Therefore , when FBF/PUF activity is lowered in the distal germline ( as germ cells leave the mitotic region and enter the transition zone ) , LIP-1/MKP abundance increases . The result of this extra step of regulation is that MAPK activity is kept low even when FBF levels decrease . Therefore , MAPK inhibition is ensured not only by redundant inhibitors but also by a well-buffered circuitry . A key unanswered question is whether mammalian MAPK is also subject to homologous redundant controls . Clearly both exist in mammals: PUM2 represses both Erk2 and p38α mRNAs ( present work ) , and MKPs negatively regulate ERK2 , p38α and JNK members of the MAPK family [6] . But do they function in the same cells in a redundant fashion ? The answer to this question will require removal of both PUF and MKP proteins in vertebrate cells , which remains a challenge for the future . In the C . elegans germline , FBF is required for stem cell maintenance [15] , and MPK-1 promotes differentiation ( either oocyte maturation or apoptosis ) [18 , 19] . Although a vertebrate role for PUF proteins in stem cell maintenance remains a matter of speculation [14 , 17] , recent evidence has given this idea credence . Thus , PUM2 is enriched in germline stem cells and embryonic stem cells [28] , and murine PUM2 mutant testes are smaller than normal and contain some agametic seminiferous tubules [32] . Therefore , the role of PUF proteins in stem cell maintenance may be conserved . The roles of MAPK and MKP in vertebrates are reminiscent of those of MPK-1 and LIP-1 in the C . elegans germline . Vertebrate ERK2 and p38α MAPKs can antagonize stem cell self-renewal and promote differentiation [9 , 33–35] . In cancer cells , ERK2 and p38α MAPKs are thought to promote apoptosis [36 , 37] . Indeed , MKPs are often upregulated in human cancer cells , and the MKP inhibition of MAPK activity has been suggested to be critical for human cancer progression [38 , 39] . Therefore , MAPKs and MKPs affect both continued self-renewal and tumor progression . In this work , we show that PUF RNA-binding proteins repress MAPK-encoding mRNAs in both C . elegans and hESCs . Indeed , to our knowledge , ERK2 and p38α mRNAs are the first PUM2 targets reported to date . The biological significance of this finding is not known . One simple idea is that PUF represses MAPK expression as part of a larger regulatory circuit designed to maintain stem cells in a naïve state . However , a more complete understanding will require learning the extent of PUM2 repression , the extent of MKP inhibition , and the biological readout of different levels of MAPK activity—all in the same cells . Although this more in-depth understanding remains a challenge for the future , we emphasize here that the PUF and MKP controls of MAPK signaling are broadly conserved and likely work together broadly to control stem cells and tumor progression .
All strains were maintained at 20 °C as described [40] , unless noted otherwise . We used the wild-type Bristol strain N2 as well as the following mutants: LGI: gld-1 ( q485 ) [41] , gld-2 ( q497 ) [42 , 43]; LGII: fbf-1 ( ok91 ) [15] , fbf-2 ( q738 ) [13] , gld-3 ( q730 ) [44] , nos-3 ( q650 ) [45]; LGIII: glp-1 ( q224 ) [46] , mpk-1 ( ga117 ) [2] , mpk-1 ( ga111 ) [24]; and LGIV: lip-1 ( zh15 ) [22] . In situ hybridization was carried out using the protocol described [47] , with minor modifications . Dissected adult hermaphrodite gonads were fixed ( 3% formaldehyde , 0 . 25% glutaraldehyde , 100 mM K2HPO4 [pH 7 . 2] ) for 3 h at room temperature . After washing three times with PBT solution ( 1× PBS containing 0 . 1% Tween 20 ) , gonads were treated with proteinase K ( 50 μg/ml ) for 30 min at room temperature and then refixed in the same solution for 15 min at room temperature . DNA probes were synthesized with digoxigenin-11-dUTP by repeated primer extension . Fixed gonads were incubated for 24 h at 48 °C in a solution containing the DNA probe plus 5× SSC , 50% deionized formamide , 100 μg/ml herring sperm DNA , 50 μg/ml heparin , and 0 . 1% Tween 20 . To visualize the probes , gonads were incubated with alkaline phosphatase−conjugated antidigoxigenin antibody ( Roche , 1:2 , 000 dilution in PBT containing 0 . 1% BSA ) at 4 °C for overnight . After washing several times in PBT ( + 0 . 1% BSA ) , staining was developed for 1 h in a solution ( 100 mM Tris Cl [pH 9 . 5] , 100 mM NaCl , 5 mM MgCl2 , 0 . 1% Tween 20 , 1 mM Levamosole ) containing 4-nitro blue tetrazolium chloride ( 0 . 23 mg/ml ) and 5-bromo-4-chloro-3-indolyl-phosphatase ( 0 . 18 mg/ml ) and then terminated in PBT containing 20 mM EDTA . Dissected gonads were fixed with 3% formaldehyde , 100 mM K2HPO4 ( pH 7 . 2 ) for 1 h , and postfixed with cold ( −20 °C ) 100% methanol for 5 min . Antibody incubations and washes were performed as described [47] . Polyclonal rabbit α-MAPK/ERK antibody ( Sc94; Santa Cruz Biotechnology ) was used at 1:400 dilution , and monoclonal mouse α-DP-MAPK antibody ( Sigma ) was used at 1:200 dilution . DAPI staining followed standard methods . Blots were prepared by standard procedures . Protein samples were separated on 4%–20% gradient gels ( Cambrex ) , and the blot was probed with polyclonal rabbit α-MAPK/ERK antibody ( Sc94; Santa Cruz Biotechnology ) , α-GFP antibody ( Molecular Probes ) , monoclonal mouse α-tubulin antibody ( Sigma ) , α-actin antibody ( MP Biomedicals ) , and α-FLAG antibody ( Sigma ) . Three-hybrid assays were performed as described [48] . For β-galactosidase assays , cells were grown in selective media to an OD600 of 1 . 0 and mixed with an equal volume of β-Glo ( Promega ) reagent . Luminescence was measured after 1 h . Gel shift assays were performed as described [49] . SYTO 12 ( Molecular Probes ) dye was used to estimate the relative numbers of germ cell corpses [19 , 50] . Animals were incubated in a 33 μM aqueous solution of SYTO 12 for 2 h at 20 °C and then transferred to seeded plates to purge stained bacteria from the intestine . After 30 min , animals were mounted on agarose pads and observed under a fluorescence microscope equipped with Nomarski optics to score SYTO 12−positive germ cells . ORF and 3′ sequences from the fbf-1 genomic locus ( from ATG to 317 bp downstream of the STOP codon ) were PCR amplified with flanking attB1 and attB2 sequences and cloned into pDONR201 ( Invitrogen ) to create pCM3 . 06 , which was sequence verified . A Gateway LR recombination reaction ( Invitrogen ) was performed between pCM3 . 06 ( entry ) and pCM2 . 03 ( destination ) . pCM2 . 03 is a bombardment-ready vector containing the unc-119 rescuing fragment ( used for transformant selection [51] ) , the pie-1 enhancer and promoter ( to drive expression in the germline [52] ) , GFP with three synthetic introns ( from pPD103 . 87 , A . Fire , personal communication ) , and the attR1::Gateway Cassette B::attR2 . The resulting plasmid pCM4 . 06 contains unc-119; pie-1 ( enhancer + promoter ) ::GFP::attB1::fbf-1ORF+3′UTR::attB2 . pCM4 . 06 was transformed into unc-119 ( ed3 ) worms by microparticle bombardment [51] to create line JH2012 ( genotype: unc-119 ( ed3 ) ; axIs1459 [CM4 . 06] ) . Age-synchronous adult Ppie-1::GFP-FBF-1 ( JK4091 ) and Ppie-1::GFP-TUB ( AZ224 ) transgenic animals were grown for 24 h after the L4 stage on NGM plates supplemented with concentrated OP50 . Worms were harvested by rinsing plates with M9 buffer , and worms were washed with M9 buffer until the supernatant was clear . Worms were then washed twice with buffer A ( 20 mM Tris [pH 8 . 0] , 150 mM NaCl , 10 mM EDTA [pH 8 . 0] , 1 . 5 mM DTT , 0 . 1% NP-40 , 0 . 02 mg/ml heparin ) , and worm pellets were frozen at −20 °C . Roughly 0 . 5 ml of worm pellets was used for affinity purification . Worm lysate was generated by grinding worms with a mortar and pestle under liquid nitrogen in lysis buffer ( buffer A plus 1× Complete Protease Inhibitor Cocktail [Roche] , 20 U/ml DNase I [Ambion] , 100 U/ml RNase OUT [Invitrogen] , 0 . 2 mg/ml heparin ) , followed by 30 passes with a glass dounce . The extract was then centrifuged twice at 10 , 000 g for 10 min at 4 °C to remove insoluble debris and fat . The protein concentrations of cleared extracts were determined by Bradford assay , and extracts were diluted to 10 mg/ml with lysis buffer . To limit nonspecific interactions with the affinity column , extracts were next precleared by incubation with 15 μl of Immobilized Protein A ( Pierce ) for 1 h at 4 °C . Then 1 . 6 μg of mouse α-GFP monoclonal antibody 3E6 ( Q•Biogene ) prebound to 16 μl of Immobilized Protein A ( Pierce ) was added to each precleared extract , and the resultant slurries were incubated at 4 °C for 2 h . Beads were then washed once with lysis buffer for 15 min at 4 °C and four times with wash buffer ( 20 mM Tris [pH 8 . 0] , 150 mM NaCl , 1 mM EDTA [pH 8 . 0] , 10% glycerol , 0 . 01% NP-40 , 1 mM DTT , 10 U/ml RNase OUT ) for 15 min at 4 °C . For protein analysis , beads were boiled in Laemmli buffer . To purify RNA , beads were treated with TRIzol reagent ( Invitrogen ) followed by RNeasy Mini Kit ( Qiagen ) to purify RNA , following manufacturer's instructions . Reverse transcription reactions were performed using 20 ng of Input or IP RNA in 10-μl reactions with an oligo ( dT ) primer and SuperScript III reverse transcriptase ( Invitrogen ) . PCR was carried out on the cDNA template for eft-3 , gld-1 , and mpk-1 in the linear range ( 33 cycles ) using gene-specific primers such that one primer spanned an exon-exon junction . PCR products were resolved on 1% agarose gel and stained with ethidium bromide . H9 human ES cells with a stably transfected EBNA protein ( H9-EBNA ) that enables episomal replication of exogenous plasmid containing an Orip site were maintained in defined medium-TeSR medium [53 , 54] containing 50 ng/ml G418 . For the transfection , H9-EBNA cells were dissociated by Dispase ( Invitrogen ) and seeded onto Matrigel-coated six-well plates . At 24 h after seeding , H9-EBNA cells were transfected with 1 μg of indicated reporter plasmids together with an equal amount of indicated effector plasmids by using Fugene6 reagents ( Roche ) . At 24 h after transfection , the cells were photographed and lysed by RIPA buffer for the Western blot analysis . | The mitogen-activated protein ( MAP ) kinase ( MAPK ) enzyme is crucial for regulation of both stem cell maintenance and tumorigenesis . Two conserved controls of MAPK include its activation by RAS signaling and a kinase cascade as well as its inactivation by MAPK phosphatases ( MKPs ) . We identify a third mode of conserved MAPK regulation . We demonstrate that PUF ( for Pumilio and FBF [fem-3 binding factor] ) RNA-binding proteins repress mRNAs encoding MAPK enzymes in both the Caenorhabditis elegans germline and human embryonic stem cells . PUF proteins have emerged as conserved regulators of germline stem cells in C . elegans , Drosophila , and probably vertebrates . Their molecular mode of action relies on binding to sequence elements in the 3′ untranslated region of target mRNAs . We report that PUF proteins bind and repress mRNAs encoding C . elegans MPK-1 as well as human ERK2 and p38α . We also report that PUF repression and MKP inactivation function redundantly in the C . elegans germline to restrict MPK-1/MAPK activity and prevent germ cell apoptosis . We suggest that this dual regulation of MAPK activity by PUF and MKP proteins may be a conserved mechanism for the control of growth and differentiation during animal development and tissue homeostasis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"cell",
"biology",
"molecular",
"biology",
"caenorhabditis",
"homo",
"(human)"
] | 2007 | Conserved Regulation of MAP Kinase Expression by PUF RNA-Binding Proteins |
Accurate and accessible diagnosis is key for the control of visceral leishmaniasis ( VL ) . Yet , current diagnostic tests for VL have severe limitations: they are invasive or not suitable as point of care ( POC ) test or their performance is suboptimal in East Africa . We analysed the antigens in the VL serodiagnostics development pipeline to identify shortcomings and to propose strategies in the development of an alternative POC test for VL in East Africa . The objective of this study was to identify and to analyse all antigens for VL serodiagnosis that have been published before 2018 in order to identify candidates and gaps in the pipeline for a new POC test in East Africa . A systematic literature search was performed on PubMed for original research articles on Leishmania-specific antigens for antibody detection of VL in humans . From each article , the following information was extracted: the antigen name , test format and characteristics , its reported sensitivity and specificity and study cohort specifications . One hundred and seven articles containing information about 96 tests based on 89 different antigens were included in this study . Eighty six of these tests , comprising 80 antigens , were evaluated in phase I and II studies only . Only 20 antigens , all of which are native , contain a carbohydrate and/or lipid moiety . Twenty-four antigens , of which 7 are non-native , are composed of antigen mixtures . Nineteen tests , comprising 18 antigens , have been evaluated on East African specimens , of which only 2 ( rK28 based immunochromatographic test and intact promastigote based indirect fluorescent antibody technique ) consistently showed sensitivities above 94 and specificities above 97% in a phase III study and one in a phase II study ( dot blot with SLA ) . Only rK28 is a non-native mixture of antigens which we consider suitable for further evaluation and implementation . The development pipeline for an alternative serodiagnostic test for VL is almost empty . Most antigens are not sufficiently evaluated . Non-protein antigens and antigen mixtures are being neglected . We propose to expand the evaluation of existing antigen candidates and to investigate the diagnostic potential of defined non-native carbohydrate and lipid antigens for VL serodiagnosis in East Africa .
Leishmaniasis comprises a group of diseases caused by protozoan parasites of the genus Leishmania that are transmitted between human or other mammalian hosts by phlebotomine sand flies [1] . Visceral leishmaniasis ( VL ) is the most severe form of the disease , for it is almost always fatal if left untreated . In 2016 , VL was reported in 21646 cases worldwide , while the estimated number of underreported cases is high ( up to 85% in 2012 ) [2 , 3] . The typical causative agent of VL is the Leishmania ( L . ) donovani complex , which includes the two species L . donovani and L . infantum . On the Indian subcontinent and in East Africa , L . donovani is almost exclusively the cause of VL[4 , 5] . Anti-leishmaniasis drugs are toxic , expensive and/or prone to induce drug resistance . The gold standard for diagnosis of VL remains the microscopic detection of the parasite in patient tissues . The most sensitive technique requires a splenic puncture , while lymph node and bone marrow aspirates can also be used with lower sensitivity [5 , 6] . These invasive diagnostics have several disadvantages , as they require highly skilled medical personnel as well as suitable clinical facilities and equipment . Furthermore , splenic and bone marrow aspirates are painful and possibly dangerous . Less invasive techniques applicable on blood exist . Due to their labour-intensiveness and their technical requirements , molecular diagnostic tests as well as other laboratory tests , such as ELISA , are not applicable in the field [5] . The ideal diagnostic test for limited resource settings should comply with the ASSURED criteria: be accurate , sensitive , specific , user friendly , rapid and robust , equipment-free , and delivered to those who need it[7] . In this regard the immunochromatographic rapid diagnostic test ( RDT ) with rK39 antigen was a major breakthrough in terms of access to VL diagnosis for populations at risk [8] . It has become the reference test for VL on the Indian subcontinent . It even has potential use to monitor treatment outcome when used to detect IgG1 [9] . The rK39 consists of 6 , 4 repeats of a 39-amino acids stretch belonging to a kinesin-related protein of L . chagasi expressed in Escherichia ( E . ) coli [8] . Unfortunately , the diagnostic accuracy of rK39-based RDTs has been shown to be dependent on the geographic region . While high sensitivities have been reported in Indian populations ( 97 . 0%; 95% CI 90 . 0 to 99 . 5 ) , reported sensitivities in East African populations are variable and generally lower ( 85 . 3%; 95% CI 74 . 5 to 93 . 2 ) [5 , 6 , 10–12] . In 2010 , an RDT with rK28 as antigen has been developed to overcome the issue of low sensitivity of rK39-based tests in Eastern Africa . rK28 is a chimeric antigen composed of three 14-amino acid repeats of the L . donovani HASPB1 gene , two 39 amino acid repeats of the L . donovani K39 kinesin protein gene and the complete open reading frame of L . donovani hydrophilic acylated surface protein B2 ( HASPB2 ) gene expressed in E . coli [13] . RDTs with rK28 antigen have been evaluated in four studies in Sudan and Ethiopia with promising results for the test manufactured by CTK Biotech ( sensitivities between 94 and 98% , and specificities between 95 and 98% ) while sensitivities and specificities based on RDTs by other manufacturers and in ELISA were more variable ( sensitivities between 89 and 100% , and specificities between 81 and 99% ) [13–17] . With rk28 having been evaluated on a relatively small number of specimens ( VL suspected patients: 285 ) , there is a risk that its performance turns out to be variable based on the population it is evaluated on . For this reason we find it important that alternative antigens are investigated . We analysed the antigens in the pipeline , to identify candidate antigens , potential gaps in the development process and to propose alternative strategies towards improved serodiagnostics for VL in East Africa . Our objective was to analyse all the published studies on serodiagnostic antigens for VL , regarding their characteristics , the test format , the study design , the number and geographic origin of tested specimens and the reported diagnostic accuracy .
A systematic literature search was performed on PubMed on 21 March 2018 of all articles published until 2018 , using the string “ ( ( ( ( diagnostic ) AND antigen ) AND visceral ) AND leishmaniasis ) NOT vaccine” . As we noticed that this search string was missing articles describing antigens which were analysed as both a vaccine candidate and diagnostic antigen a next screen was performed with the search string " ( ( ( ( diagnostic ) AND antigen ) AND visceral ) AND leishmaniasis ) AND vaccine" . Results of both search strings were merged and analysed together . Titles and abstracts were screened by two independent investigators ( VK&ZR ) . In case of non-concordance , a third independent investigator ( PP ) screened title and abstract of the respective articles . Both the second and third investigator were blinded to the other investigators’ decisions . If the first investigator contested the second and third investigators decision , a senior investigator ( PB ) was consulted for a final decision . There was no formal study protocol and the study was not registered . Based on information in the title and abstract , all original research articles that reported on the diagnostic potential of one or more molecules for antibody detection of VL in humans , were included . Systematic reviews and meta-analyses on such articles were also included . Exclusion criteria were: i ) only non-human ( e . g . canine ) specimens ( serum or plasma ) were tested , ii ) only non-VL specimens ( e . g . tegumentary , cutaneous , mucocutaneous leishmaniasis ) were tested , iii ) study on non-antibody detection tests ( e . g . molecular , microscopic , antigen detection ) , iv ) study on antigens that were not used for serodiagnosis of VL ( e . g . test for experimental infection or treatment outcome or for epidemiological surveys ) , v ) study on immunological biomarkers such as cytokines . The remaining articles were analysed based on their full text . Additional exclusion criteria were: i ) no control specimens were tested ( other disease or healthy endemic controls ) ; ii ) less than 5 specimens tested; iii ) antigen with poor diagnostic potential ( < 50% sensitivity and/or < 50% specificity ) ; iv ) data for sensitivity and specificity not presented; v ) number or geographic origin of specimens not described vi ) antigen not Leishmania-specific ( e . g . BCG , Kendrows Bacillus ) ; vii ) article written in a language other than English , French or German; viii ) full article text not available . Antigens ( rK39 and stained formaldehyde-fixed promastigotes for direct agglutination test [DAT] ) that were already included in a meta-analysis by Chappuis et al . ( 2006 ) [10] or Boelaert et al . ( 2014 ) [6] were not re-analysed but the results of the meta-analysis were taken into account [6 , 10] . For each included article , the following data were imported in a Microsoft Excel worksheet: antigen name , molecular weight , reported sensitivity and specificity , study design , number or VL cases and controls tested , type of assay , type of controls tested , type of specimen tested , geographical origin of the tested specimens , Leishmania species and strain from which the antigen was derived , method of antigen preparation/production , article reference . The individual studies were not analysed for risk of bias as we intended to analyse the complete data on antibody detectiontests for VL including early discovery studies with limited number of specimens and high risk of bias . We did not perform a meta-analysis on the diagnostic sensitivity and specificity of the tests , which reduces the potential bias of the individual studies . The data were analysed in three ways:
Data from the included articles were grouped per type of test , test antigen and study design ( phase I , II and III ) . For studies on the same test in the same phase , the total numbers of cases and controls were calculated . We found 96 serodiagnostic tests comprising 89 antigens for human VL ( Fig 2 , Table 1 ) . As Bhattacharyya et al . pointed out in their recent review , the use of "rk" in the antigen nomenclature is misleading [21] . Eight different antigens included in our analysis are based on genes for kinesin related proteins such as rk39 ( Table 2 ) . Forty four tests representing 42 antigens are evaluated on ≤100 specimens in phase I studies , 57 tests representing 53 antigens were evaluated on >100-specimens in phase II studies , 10 tests representing 9 antigens , were evaluated in phase III studies . Thus 86 out of the 96 tests ( 89% ) were evaluated in phase I and/or II only . Seven tests ( 6 antigens ) in phase III are evaluated on less than 600 specimens , while 3 antigens/tests are evaluated on 601 to 5000 specimens and no antigen is evaluated on >5000 specimens . Seventeen of the tests in phase I , 11 in phase II and 1 test in phase III ( rK28 immunochromatographic test [ICT] from CTK ) consistently have sensitivities above 94% and specificities above 97% in all studies analysed ( green in Fig 2 and grey in Table 1 ) . Fifty-three antigens ( 59% ) were expressed in E . coli or on phages or were synthetic peptides and thus do not contain eukaryotic glycosylated or lipid moieties . Two of the native antigens were analysed for glycosylation and do not contain a carbohydrate moiety . P32 and the 62 to 63 kDa protein were shown to have no glycosylation by acid-Schiff staining and the latter also does not bind Concanavalin A [30 , 68] . The remaining 34 ( 38% ) are divided in three categories . In 14 , the nature of the antigen could not be deducted from its isolation procedure ( parasite lysate run on sodium dodecyl sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) and Western blot , antigen eluted form SDS-PAGE , acid extraction , isolation using a Lipidex-1000 column [34 , 41 , 44 , 48 , 63 , 104 , 118 , 121] ) and no additional analysis has been performed to conclude on its nature . Five antigens bear a carbohydrate while it is not certain whether they bear a lipid moiety , such as the concanavalin-A specific antigen isolated by Bhattacharya [54] , gp 70–2 which reacts with the monoclonal D2 and was sensitive to periodate treatment [55] , two competition ELISAs using D2 and other monoclonal antibodies [46 , 50] and FML was found to be composed of sugars by colorimetry [36] . Fifteen antigens contain both a carbohydrate and a lipid moiety as they are made from whole parasites , whole parasite lysates , the entire secretome , the parasite membrane or lipophosphoglycan ( LPG ) [31 , 32 , 35 , 60 , 62 , 111 , 114 , 122] . The second part of the analysis of the antigen characteristics is summarized in Fig 3 . All 20 antigens with a known carbohydrate and or lipid moiety are native antigens . Fourteen are antigen mixtures of undefined identity , such as the DAT [122] and soluble leishmanial antigen ( SLA ) ( e . g . [62] ) which consist of a mixture of parasite molecules including carbohydrates and lipids . Two are antigen mixtures which can be considered as defined: FML [36 , 37] and the competition ELISA using the monoclonal antibodies D2 , D13 and D14 to inhibit antibodies reacting with SLA [50] . Four antigens containing a carbohydrate and/or lipid moiety can be considered defined: lipophosphoglycan [31] , concanavalin-A specific antigen [54] , gp 70–2 [123] and the inhibition ELISA using the monoclonal antibody D2 ( which is directed against gp 70–2 ) to inhibit antibodies reacting with SLA [46] . Among the antigen mixtures that lack a known carbohydrate or lipid moiety , 8 are defined . One is a native antigen ( crude Leishmania histone ( CLH ) [48] ) while the other 7 are non-native . Nineteen tests , representing 19 ( 21% ) antigens , were evaluated on East African specimens . Fig 4 and Table 3 show the position of these tests and antigens along the diagnostic development pipeline . Only Indirect Fluorescent Antibody Technique ( IFAT ) and the rK28 ICT were shown to be >94% sensitive and >97% specific when evaluated in phase III studies on respectively 104 and 285 East African persons [15 , 108] . The other tests in phase III are rK39 ICT , rKE16 ICT , beta-mercaptoethanol modified ELISA ( ß-ME ELISA ) , DAT and FAST . They were evaluated on respectively 1692 , 219 , 490 , 208 and 148 persons but showed lower sensitivities and/or specificities in at least one of the studies [6 , 10 , 29 , 32 , 33] . All the other tests and antigens have only been evaluated in phase I and II studies . Among those , the SLA in dot blot ( evaluated on 126 specimens ) and the 70 kDa soluble antigen in ELISA ( evaluated on 42 specimens ) , were >94% sensitive and >97% specific [38 , 106] .
There is a multitude of publications on new diagnostic tests for VL , of which many report very high diagnostic accuracies ( sensitivities and specificities up to 100% ) . While this seems very promising , diagnostic accuracies of tests are difficult to be compared to one another as they are not necessarily assessed on a comparable number and type of specimens . We analysed the level of evaluation and the diagnostic accuracies reported for every test published until 2018 to situate them in the diagnostic development pipeline and to facilitate comparisons between tests and antigens used in these tests . According to Boelaert et al . 2007 [18] the diagnostic development pipeline comprises three major phases defined by purpose and study design . Phase I and phase II follow a case-control design and deliver the proof-of-principle and intrinsic sensitivity and specificity of a diagnostic test , which are mostly defined by the antigen and the assay format . To validate the diagnostic accuracy of a test , it should undergo an evaluation in a prospective phase III study conducted on a larger number of representative samples of consecutively enrolled or randomly selected patients within the population [18 , 19] . Our analysis shows that 89% of the tests were evaluated in phase I and/or phase II . To a certain extent , this can be explained by the fact that underperforming candidates are not taken further along the development process . This is however not the case for 13 of the antigens in phase I and 10 of the antigens in phase II with observed sensitivities and specificities higher than 94 and 97% respectively that were not evaluated in the following phase . Why these antigens were not taken to the next level of evaluation cannot be explained by our data . It does however fit with the often described “valley of death” between discovery and commercialisation [124] , a term that is used to describe the deficiencies in translating discoveries from ( mostly academic ) research into commercial products by the industry . There seems to be a lack of connectivity between the different stages of development , which hampers the advance of new diagnostics tools , leaving promising compounds on the back shelf . Moreover , it creates the impression of a vast variety diagnostic candidates , while they were never sufficiently evaluated to make conclusions on their performance . The number of extensively evaluated tests with high diagnostic sensitivity and specificity is thus very low , which can be explained by the correlation between the number of specimens and studies in which a test has been evaluated and its probability to be screened out for its insufficient diagnostic accuracy . In a recent literature review on the nature of the antigen in the DAT , which is highly sensitive and specific for VL diagnosis across its geographical distribution , we concluded that alternative candidate antigens for VL serodiagnosis should be composed of a mixture of antigens that also contains a non-protein moiety and that can be produced recombinantly or synthetically [125] . From the present analysis , it appears that no antigen candidate with these characteristics exists in the diagnostic development pipeline for VL diagnosis . Fifty three antigens are recombinantly expressed in E . coli or on phages or are synthetic peptides . These proteins and peptides do not undergo eukaryotic posttranslational modifications like glycosylation . Thus , epitopes based on carbohydrate and lipid moieties present on the original parasite components are not represented on these antigens . Of the remaining 36 antigens , only 20 certainly bear a carbohydrate and/or lipid moiety . Among the 96 diagnostic tests analysed , we were able to identify 24 antigens in the diagnostic development pipeline that contain antigen or epitope mixtures . Ideally , an alternative serodiagnostic test for VL should be manufactured at low cost and with a high level of standardisation . Therefore , using native antigens is less desirable since their production is usually expensive and inherently prone to considerable batch to batch variation . Even more so when the antigens are not defined in the mix of molecules used . We therefore analysed how many of the carbohydrate and / or lipid containing antigens and how many of the antigen mixtures are native and whether they are composed of more or less defined molecules , as opposed to being crude parasite extracts or whole cell antigens . We found that all 20 antigens that contain a non-protein moiety are native and 14 are undefined . Sixteen antigens with carbohydrate and/or lipid moieties are antigen mixtures , 14 of which are undefined . Only 7 out of the 24 antigen mixtures are not native , i . e . consist of recombinants or peptides . Considering that a reliable point of care ( POC ) test for VL in East Africa is not secured yet , we analysed the tests in the diagnostic development pipeline that have been tested on sera from East African VL patients . We found that only the IFAT and the rK28 ICT [15 , 108] have shown a high diagnostic accuracy ( sensitivity and specificity above 94 and 97% respectively ) in cohorts of VL suspected patients in East Africa . All other candidates with high diagnostic accuracies on East African specimens ( dot blot with SLA [106] and a ELISA with a 70 kDa soluble antigen [38] ) were evaluated in case-control designed studies and might be screened out upon further evaluation . There is 1 non-native , defined antigen with high diagnostic accuracy on East African specimens: the rK28 , which incorporated in an ICT has been evaluated on 285 persons in phase III . Apart from this promising candidate the pipeline of antibody detection tests for VL detection in East Africa almost empty . Our analysis was limited by the fact that we only searched one database ( PubMed ) . The data extraction was performed by one investigator only , which might have led to errors . We did not consider the risk of bias on individual study level which might have impacted the analysis on the diagnostic sensitivities and specificities of the tests . However our aim was to show the landscape of existing tests and tendencies within it , which we believe is not hindered by the limitations of the study , as the search and analysis were performed systematically . With our analysis on the development pipeline of new serodiagnostic POC tests for VL especially for East Africa , we identified five major problems: ( i ) The number of tests and antigens that have been extensively evaluated in phase III studies is low , with more than three quarters of all tests having been evaluated in phase I and II . Especially in Eastern Africa where antigen diversity is high compared to South-Asia [126] we consider it essential to conduct a large-scale evaluation on a broad panel of sera from different geographic origin , before concluding on the diagnostic potential of a test . We recommend including a sufficient number of specimens from Eastern Africa in the downstream evaluation of any new POC test for VL showing high diagnostic accuracy in early phases of development . A major obstacle for this is the limited number of and access barriers to reference serum archives ( ii ) Not enough emphasis is put on non-protein moieties of candidate antigens , with only 20 of the 89 antigens bearing a defined carbohydrate or lipid moiety . ( iii ) The fact that all antigens with carbohydrate and/or lipid moieties and 17 of the 24 mixed components antigens are native , poses a problem for the eventual commercialisation in terms of production cost and batch-to-batch variation . For well identified antigens or antigen mixtures , eukaryotic expression systems such as developed in Leishmania tarentolae , Pichia pastoris or Spodoptera frugiperda should be envisioned , as they allow to produce recombinant proteins with eukaryotic post-translational modifications , including glycosylation [127–129] . For undefined antigens , screening of peptide libraries could be considered to select short peptides ( mimotopes ) that mimic native carbohydrate and lipid epitopes . Umair and co-workers proved this concept by using phage display technology to discover a mimotope that can replace the glycan epitope on the surface of parasitic nematode larvae [130] . ( iv ) Mixtures of antigens are not sufficiently taken into account; only 7 candidate recombinant antigens consist of a mixture of several compounds . The development of non-native antigen mixtures should be extended . We see two possibilities for creating antigen mixtures: a ) chimeric recombinant antigens such as rk28 [13] , Q protein [73] , the chimeric antigen with epitopes of L . infantum K9 , K26 , and K39 [59] and the synthetic peptide P1P2 which consists of two epitopes of the hypothetical protein NCBI reference: XP_003861458 , 1 [26] b ) combinations of several antigens such as the mixture of synthetic peptides derived from gp63 on human serum albumin [71] , of synthetic peptides from antigens A2 , NH , LACK and K39 [69] and the OrangeLife ICT which is based on an association of rK39 and rK28 [22] . ( v ) 70% of all carbohydrate and lipid containing antigens and 71% of all antigen mixtures are non-defined . Aiming at the development of alternative diagnostic test for VL in East Africa that complies with the ASSURED criteria , investigations should be undertaken to identify these reactive epitopes and to replace them by recombinant or synthetic antigens . | Visceral leishmaniasis is a potentially fatal disease that affects more than 20 000 people every year . Its diagnosis is difficult since the clinical symptoms are not specific and the existing diagnostic tests are not useful in limited resource countries or they a not accurate enough in East Africa . In this review we performed a systematic search of the published literature to analyse the potential candidate antigens in the pipeline for a new antibody detection test in East Africa . We found 96 tests based on 89 antigens . Eighty six of these tests were evaluated in a study design that is insufficient ( phase I and II ) to make conclusions on their performance in clinical practice . We found that the candidate antigens either lacked carbohydrate or lipid structures or are based on single antigens as opposed to mixtures or are extracted from the causative parasite itself , making them expensive and prone to variations . Considering that the most widely used diagnostic test does not detect all cases of visceral leishmaniasis in East Africa , we analysed how many of the candidate antigens were tested on East African specimens: We found that only 2 tests ( rK28 based immunochromatographic test and the intact promastigote based indirect fluorescent antibody technique ) that were tested in a phase III study and only one ( dot blot with SLA ) that was tested in a phase II study performed well according to our criteria . Due to the antigen characteristics we consider only the rK28 based test as suitable for further evaluation and implementation . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"chemical",
"compounds",
"immunology",
"tropical",
"diseases",
"microbiology",
"carbohydrates",
"parasitic",
"diseases",
"organic",
"compounds",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"developmental",
"biology",
"protozoans",
"leishmania",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"promastigotes",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"lipids",
"zoonoses",
"proteins",
"antigens",
"immunoassays",
"life",
"cycles",
"chemistry",
"protozoan",
"infections",
"leishmania",
"donovani",
"biochemistry",
"eukaryota",
"organic",
"chemistry",
"physiology",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"protozoology",
"physical",
"sciences",
"organisms"
] | 2019 | Systematic review on antigens for serodiagnosis of visceral leishmaniasis, with a focus on East Africa |
Schistosomiasis remains an important public health issue in China and worldwide . Oncomelania hupensis is the unique intermediate host of schistosoma japonicum , and its change influences the distribution of S . japonica . The Three Gorges Dam ( TGD ) has substantially changed the ecology and environment in the Dongting Lake region . This study investigated the impact of water level and elevation on the survival and habitat of the snails . Data were collected for 16 bottomlands around 4 hydrological stations , which included water , density of living snails ( form the Anxiang Station for Schistosomiasis Control ) and elevation ( from Google Earth ) . Based on the elevation , sixteen bottomlands were divided into 3 groups . ARIMA models were built to predict the density of living snails in different elevation areas . Before closure of TGD , 7 out of 9 years had a water level beyond the warning level at least once at Anxiang hydrological station , compared with only 3 out of 10 years after closure of TGD . There were two severe droughts that happened in 2006 and 2011 , with much fewer number of flooding per year compared with other study years . Overall , there was a correlation between water level changing and density of living snails variation in all the elevations areas . The density of living snails in all elevations areas was decreasing after the TGD was built . The relationship between number of flooding per year and the density of living snails was more pronounced in the medium and high elevation areas; the density of living snails kept decreasing from 2003 to 2014 . In low elevation area however , the density of living snails decreased after 2003 first and turned to increase after 2011 . Our ARIMA prediction models indicated that the snails would not disappear in the Dongting Lake region in the next 7 years . In the low elevation area , the density of living snails would increase slightly , and then stabilize after the year 2017 . In the medium elevation region , the change of the density of living snails would be more obvious and would increase till the year 2020 . In the high elevation area , the density of living snails would remain stable after the year 2015 . The TGD influenced water levels and reduced the risk of flooding and the density of living snails in the study region . Based on our prediction models , the density of living snails in all elevations tends to be stabilized . Control of S . japonica would continue to be an important task in the study area in the coming decade .
Schistosomiasis remains a serious public health problem worldwide , affecting more than 200 million people in approximately 76 countries with a loss of 1 . 53 million disability-adjusted life years ( DALYs ) [1] . Schistosomiasis japonica ( S . japonica ) is distributed in 12 provinces in China and 11 . 6 million people have been infected since 1949 [2–4] . S . japonica causes the most hazardous schistosomiasis , and is difficult to prevent and treat [5 , 6] . After continued implementations of comprehensive control measures from the mid-1950s to 1980s , endemic regions were circumscribed in certain core areas in China , especially in the Dongting Lake region at the middle reaches of the Yangtze River [7–11] . Most of these core areas are in the downstream of Three Gorges Dam ( TGD ) . The TGD is a world-class water conservancy project . It began to impound water and sediment discharge in 2003 , and is one of several tremendous engineering projects transforming China’s ecology and natural environment . The construction and the operation of TGD have obviously affected the ecological environment . The water level of TGD in 2003 was 135m and it reached 160m in 2011 . It is believed that the TGD project reduces the frequency of major flooding in the downstream areas from once every ten years to once every 100 years; however it threatens the living of aquatic animals in the Yangtze River including the river dolphin , or baiji , and finless porpoise , or jiangzhu [12] . Similarly the TGD project can also influence the survival of O . hupensis , the unique intermediate host of Schistosoma japonicum [13 , 14] . The TGD has been completed for over 10 years , and its impact on the transmission of S . japonica has to be evaluated [15 , 16] . This study aimed to determine the impact of changes in water level and number of flooding per year on the snail ( O . hupensis ) density , and to predict the changes in the density of living snails in the Dongting Lake region using the autoregressive integrated moving average ( ARIMA ) model , which combines the advantages of autoregressive ( AR ) model and moving average ( MA ) model .
This study was conducted in the 16 bottomlands near four hydrological stations in Anxiang County of the Dongting Lake region ( Fig 1 ) . These bottomlands are outside the embankment . The Dongting Lake is located at 28°30′–30°20′N and 111°40′–113°40′E in the northeastern part of Hunan Province and covers a water surface area of 2 , 681 km2 , and plays an important role in regulating the amount of water in the Yangtze River [17] . It is a typical S . japonica endemic area with an at-risk population of about 429 , 000 people in marshland and lake regions [18] . Anxiang County is located in the Dongting Lake area and at the downstream of the TGD . There are 4 hydrological stations located in Anxiang County and for each hydrological station we selected 4 bottomlands as our study fields ( Fig 1 ) : 1 ) Guanyuan hydrological station—Qinglong , Dongbao , Nanyang and Yangshutan; 2 ) Zizhiju hydrological station—Yandoukou , Dongdi , Yongheyuan , Yifenju; 3 ) Shiguishan hydrological station—Qiangkou , Huangjiatai , Fuxing , Wuyang; and 4 ) Anxiang hydrological station—Yucheng , Liujiao , Zhulin , Anxiang . A total of sixteen bottomlands were divided into 3 groups according to three elevation: Low ( <33m ) , medium ( 33 to 35m ) and high ( >35 m ) . The routine data of snails which could be obtained and used was snail density and snail area . The change in snail areas was not large in this study field and period . The density of living snails is an important index reflecting the survival and reproduction of snails . Thus the density of living snails was used to assess the impact of TGD on snails . The Anxiang Station for Schistosomiasis Control provided data of density of living snails in every bottomland of Anxiang County for the period from 2004 to 2014 . Snail surveys were conducted in each spring and were implemented using a traditional method of random quadrant sampling ( 0 . 11 m2-sized frames , 20 m apart between frames ) [19] . We downloaded data of water levels from the Hunan Flood Prevention Information System , including daily water levels at 8:00 am at the 4 hydrological stations from 1995 to 2013 . The elevation for the 4 hydrological stations was 33m for the Guanyuan station , 31m for the Zizhiju and Anxiang stations , and 30m for the Shiguishan station . Data of elevation were collected through the Google Earth ( Google Ltd , USA , http://www . google . com/earth/ ) by sampling 50 points ( 5*10 , 200 m apart between points ) for each bottomland ( Fig 2 ) . A weighted average elevation was calculated for each bottomland . We calculated number of flooding per year , which is calculated in the basis of the difference between the daily water level at each hydrological station and the weighted average elevation of the bottomland near the station . If the difference was greater than 0 , the bottomland around the hydrological station was considered to be flooded . We used the density of living snails time-series data to fit an optimum ARIMA model and predict the density of living snails in the future . ARIMA model is a component of the Box-Jenkins approach to time-series modeling [20–22] . An ARIMA model is derived by combining the three techniques: autoregressive modeling ( AR ) , moving average modeling ( MA ) and differencing . They are presented as ARIMA ( p , d , q ) , in which p and q represent the orders of AR and MA models , respectively , and d denotes the order of differencing . In ARIMA models , we assume a stationary time series , which means that the data vary around a constant mean and variance over time [23–25] . Nonstationary time series variables can be converted into stationary ones . An ARIMA ( p , d , q ) model can be written as [26 , 27]: y't=c+ϕ1y’t-1+…+ϕpy’t-p-θ1zt-1-…-θ1zt-1+zt where c is a constant , y’t = yt-yt-1 represents the differenced series , y’t-p are lagged value and zt is a white noise process . The ARIMA modeling procedure consists of three iterative steps: selecting a candidate model , estimating the model and performing diagnostic tests and forecasting . Before fitting an ARIMA model , selecting a candidate model is the process of identifying randomness , stationarity and seasonality using the autocorrelation functions ( ACF ) and partial autocorrelation functions ( PACF ) . If data do not meet these requirements , a transformation of data should be implemented [28] . ACF is a statistical tool that measures whether an earlier value in the series has some relation to a later value . PACF captures the amount of correlation between a variable and a lag of this variable that is not explained by correlation at all low-order lags . Parameters in the ARIMA model ( s ) are estimated with the conditional least squares ( CLS ) method [29] . Since there are three parameters in an ARIMA model , different parameter combinations will lead to various results . The second step is to fit an optimum ARIMA model based on the Bayesian information criterion ( BIC ) , for which the less BIC is , the better the model fits the data . Finally , the fitted model can be used to forecast the density of living snails and its confidential interval [30] . IBM SPSS 20 . 0 ( IBM Corporation USA , http://www-01 . ibm . com/software/analytics/spss/ ) was used for all the analysis .
Fig 3 shows the water levels at the Anxiang hydrological station from 1995 to 2013 , with a warning water level for flooding of 37 . 0 m . June 1st 2003 is the day that TGD began to impound . Before the year of 2003 , 7 out of 9 years had a water level beyond the warning level at least once at this station compared with only 3 out of 10 years after the year of 2003 ( 2003 not included . Fig 3 also shows similar decreasing trends for the occurrence frequency of flooding and water level after 2003 . Fig 4 shows the number of flooding per year and density of living snails in low , medium and high elevation areas . The data of 2009 were not included due to its too many outliers . The density of living snails was correlated with the water level a year before , which is caused by a lag effect of the water level . There were two severe droughts happened in 2006 and 2011 . The density of living snails in the high and medium elevation areas had a continuously decreasing trend from 2004 to 2014 . In the low elevation area , the density of living snails had a similar trend between 2004 and 2012 but turned to increase after 2012 . In the medium elevation areas , the density of living snails decreased sharply from 2004 to 2008 and stabilized after 2008 . Table 1 presents number of flooding per year before and after 2003 in 3 different elevations . After the construction of TGD , the number of flooding significantly decreased in the areas of both medium elevation ( t = 4 . 519 , p<0 . 000 ) and high elevation areas ( t = 3 . 475 , p = 0 . 001 ) , but increased slightly in the low elevation area ( t = -1 . 428 , p = 0 . 158 ) . ARIMA model was fitted to predict the density of living snails in the future in the 3 different elevation areas in the basis of historical time-series data . Autocorrelation graphs of the density of living snails indicated that the data were randomized and smooth . The parameter group ( 1 , 0 , 0 ) was preferred according to the BIC values of the density of living snails in the low , medium and high elevation areas ( -4 . 12 , -0 . 61 and -2 . 71 ) . Fig 5 shows the results of ARIMA ( 1 , 0 , 0 ) model in the low elevation area , which generates predicted values based on historical time-series data . The predicted value of the density of living snails in the low elevation area would increase from 0 . 37 /0 . 11 m2 in 2015 to 0 . 40/0 . 11 m2 in 2019 and after . Similarly , Fig 6 shows the results of ARIMA ( 1 , 0 , 0 ) model for the medium elevation area , and the predicted value of the density of living snails would increase from 0 . 42/0 . 11m2 in 2015 to 0 . 71/0 . 11m2 in 2020 . Fig 7 presents the findings of ARIMA ( 1 , 0 , 0 ) model for the high elevation area , and the predicted values would not change markedly between 2015 and 2020 .
After the impoundment of TGD in 2003 , the dry season in Dongting Lake region was reported to arrive earlier and to be longer than before , and the water level was close to the lowest level in history for several times [31–34] . Anxiang County of Hunan Province is located in Dongting Lake region and at the downstream of TGD . Our results indicated that the water levels in all the elevation areas and number of flooding decreased after 2003 and that there was a severe drought in 2006 . Overall , there was a correlation between water level changing and density of living snails variation in all the elevations areas . The density of living snails in all elevations areas was decreasing after the TGD was built . The relationship between number of flooding per year and the density of living snails was more complicated . In the medium and high elevation areas , the density of living snails kept decreasing from 2003 to 2014 . In low elevation area , however , the density of living snails decreased after 2003 first and turned to increase after 2011 . The data of 2014 show that , the value of the density of living snails high in the low elevation area , low in the high elevation area and in between in the medium elevation area . Number of flooding per year decreased after 2003 , so was the density of snails in the medium and high elevation areas . The association did not reach statistical significance in the low elevation area . Oncomelania is an amphibious snail , and its larva needs to live in water . When it grows into adult stage , it tends to inhabit in a humid region , like grass . Water is one of the necessary conditions for growth and reproduction of the snails that have to live in water or a wet place , and it is difficult to survive in the dry environment [35–37] . In the medium and high elevation areas , the reduced number of flooding per year might result in droughts in some months , which led to the decreased density of living snails . It might not be the case in the low elevation area . It is reported that the density of living snails in middle reaches of the Yangtze River including Dongting Lake Region is sharply reduced from 2003 to 2014 [38] . Another research finds that the density of living snails in Poyang Lake Region is declined after the impoundment of TGD [39] . The results in this study are consistent with the previous findings . Droughts in this area are associated with the density of living snails . The severe drought in 2006 caused a decline of the density of living snails in 2007 . The frequency of droughts in the medium and high elevation areas after 2003 was more than that before 2003 , and as a consequence the density of living snails had a marked change . In the low elevation area however , the frequency of droughts after 2003 was less than that before 2003 and the decline of the density of living snails in this region was less obvious as compared with other areas . Based on our prediction models , the snails would not disappear in the Dongting Lake region in near future . In the low elevation area , the density of living snails would increase slightly and then stabilize after the year of 2017 . In the medium elevation region , the change of the density of living snails would be more obvious and would increase till the year of 2020 . In the high elevation area , the density of living snails would remain stable after the year of 2015 . Control of the snails would continue to be a challenge in the study area in the coming decade , and some measures , such as snail surveillance and fencing bovines on the marshland , need to continue to be implemented . There are some limitations in this study . Firstly , data of the density of living snails were not available before 2003 and we were not able to make a comparison of the density of living snails before and after TGD was built . Secondly , only water level and elevation were studied while other factors such as climate conditions and plant types were not considered . However , we did not identify any important factors that had changed dramatically during the study period . In conclusion , hydrology is an important determinant for the density of living snails . The density of living snails changed in different elevation areas , which was correlated with the variation of hydrology . TGD influenced the hydrology in the study areas and further changed the density of living snails . Based on the results from ARIMA models , we predicted that controlling snails would continue to be a challenge in the study area in the coming decade although TGD might lead to a reduction in the density of living snails in the region . | Oncomelania hupensis , an amphibious animal , is the unique intermediate host of schistosoma japonicum . Three Gorges Dam ( TGD ) is a tremendous hydrological project , and it influences the survival of animals downstream . It is studied for several reasons . First , schistosomiasis is still a world-wide parasitic disease which needs to be prevented and controlled . Second , TGD causes the change of water level , and it will impact the existence of snails , but how TGD causes the change of snails and even the epidemics of schistosomiasis is not revealed . In this study , the authors explore the association between TGD and snails . The time-series data contains over 10-year water level downstream TGD and 10-year density of living snails downstream TGD in Dongting Lake Region . These can help to find out the relationship between the change of water level and the change of density of living snails . After this exploration , we attempt to predict the density of living snails 7years later using ARIMA model . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Three Gorges Dam: Impact of Water Level Changes on the Density of Schistosome-Transmitting Snail Oncomelania hupensis in Dongting Lake Area, China |
The MORC family of GHKL ATPases are an enigmatic class of proteins with diverse chromatin related functions . In Arabidopsis , AtMORC1 , AtMORC2 , and AtMORC6 act together in heterodimeric complexes to mediate transcriptional silencing of methylated DNA elements . Here , we studied Arabidopsis AtMORC4 and AtMORC7 . We found that , in contrast to AtMORC1 , 2 , 6 , they act to suppress a wide set of non-methylated protein-coding genes that are enriched for those involved in pathogen response . Furthermore , atmorc4 atmorc7 double mutants show a pathogen response phenotype . We found that AtMORC4 and AtMORC7 form homomeric complexes in vivo and are concentrated in discrete nuclear bodies adjacent to chromocenters . Analysis of an atmorc1 , 2 , 4 , 5 , 6 , 7 hextuple mutant demonstrates that transcriptional de-repression is largely uncoupled from changes in DNA methylation in plants devoid of MORC function . However , we also uncover a requirement for MORC in both DNA methylation and silencing at a small but distinct subset of RNA-directed DNA methylation target loci . These regions are characterized by poised transcriptional potential and a low density of sites for symmetric cytosine methylation . These results provide insight into the biological function of MORC proteins in higher eukaryotes .
Maintaining regulatory access to genes while repressing the expression of potentially deleterious transposable elements is a fundamental challenge for living organisms . Eukaryotes achieve this in part by parsing their genomes into functional units characterized by distinct chromatin features [1 , 2] . The most stable chromatin mark is cytosine DNA methylation [3] . In plants , DNA methylation is often associated with transcriptionally silent regions [4 , 5] and occurs primarily in three sequence contexts , CG , CHG and CHH ( where H is defined by any base except G ) . Methylation at the symmetrical CG and CHG sites is maintained by the action of MET1—the homologue of mammalian DNMT1—and CMT3 , respectively [6] . Asymmetric CHH methylation must be continuously re-established . In pericentromeric heterochromatin , this is mostly mediated by CMT2 [7 , 8]; while in small patches of heterochromatin in the otherwise euchromatic arms , CHH methylation is mostly maintained by the action of DRM2 in the RNA-directed DNA methylation ( RdDM ) pathway [9–11] . RdDM primarily targets transposable elements through the combined action of two plant specific RNA polymerases [12 , 13] . During RdDM , Polymerase IV ( Pol IV ) is in part recruited by SHH1 [14] to generate short transcripts [15–17] , which are made double-stranded by the action of RDR2 and diced into 24nt small RNAs by DCL3 . Polymerase V ( Pol V ) is targeted to methylated sites via SUVH2/9 [18 , 19] and generates scaffold transcripts to recruit 24nt small RNA directed complexes [20 , 21] , which then recruit the de novo methyltransferase DRM2 to induce DNA methylation in all sequence contexts [10] . The RdDM pathway results in a robust self-reinforcing loop; however , a potential role for 21nt small RNAs and RDR6 during the early stages of methylation establishment has recently emerged [22–24] . To identify novel factors involved in transcriptional gene silencing , forward genetic screens from three independent laboratories isolated alleles of AtMORC6 [NP_173344; AT1G19100; CRH6; Defective in Meristem Silencing 11 ( DMS11 ) ] [25–27] . MORC proteins are members of the GHKL ATPase superfamily [28 , 29] and by evolutionary comparison with prokaryotes are predicted to play a role DNA superstructure manipulations in response to epigenetic signals [30] . While the involvement of AtMORC6 in transcriptional repression is established , the extent to which it contributes to DNA methylation at target loci has varied between reports [25–27] . For instance , a 2012 study [25] found little evidence for methylation changes at either the de-repressed reporter construct or genome wide , while Lorković et al . , 2012 [26] and Brabbs et al . , 2013 [27] both observed minor reductions in DNA methylation at their reporter loci . It therefore remains uncertain whether transcriptional activation is associated with loss of DNA methylation in atmorc mutants and to what extent AtMORC proteins are involved in the RdDM pathway . Another member of the A . thaliana MORC family , AtMORC1 [NP_568000; AT4G36290; Compromised Recognition of Turnip Crinkle Virus 1 ( CRT1 ) ] , is involved in plant defense and was isolated as a mutant that is hyper-sensitive to Turnip Crinkle Virus [31] . Interestingly , AtMORC1 was also identified in the same transcriptional repression screen that isolated AtMORC6 [25] . Recent studies have implicated changes in DNA methylation and transcriptional responses to pathogen infection [32–34] . Yet it is unclear how AtMORC1 might function in both plant defense and transcriptional repression at RdDM targets . AtMORC1 and its very close homolog AtMORC2 act in mutually exclusive heteromeric complexes with AtMORC6 , and an atmorc1 atmorc2 atmorc6 triple mutant resembles that of atmorc6 with regard to transcriptional profile and methylation state [35] . As there are seven members of the MORC family in Arabidopsis , we sought to characterize the remaining AtMORC genes in order to help elucidate MORC function . We found that the highly related AtMORC4 [NP_199891; AT5G50780; CRH4] and AtMORC7 [NP_194227; AT4G24970; CRH3] proteins act partially redundantly to transcriptionally repress a large regulon and also play a role in plant defense . Both AtMORC4 and AtMORC7 were found to form stable homomers , but do not interact with each other , suggesting that they act in parallel to control gene silencing . We also found that AtMORC4 and AtMORC7 , like AtMORC1 and AtMORC6 [25] , form nuclear bodies that are adjacent to chromocenters . Finally , by generating a compound mutant devoid of all MORC function , we demonstrate that transcriptional de-repression can be largely uncoupled from changes in DNA methylation . However , a small but distinct subset of RdDM loci that are poised for transcriptional reactivation exhibit MORC-dependent methylation changes and reduced symmetric methylation potential .
AtMORC4 and AtMORC7 are highly related to one another ( Fig 1A and 1B ) [35] . We obtained T-DNA knockout lines for these genes ( atmorc4-1 and atmorc7-1 ) ( S1A Fig ) . RT-PCR at targets known to be de-repressed in the atmorc6 background [25 , 35] showed little change in transcript levels in the homozygous knockouts . However , when we crossed the lines to create an atmorc4-1 atmorc7-1 double knockout , we observed de-repression at several of the candidate loci , suggesting that AtMORC4 and AtMORC7 act redundantly ( S1B Fig ) . To determine the extent of redundancy between AtMORC4 and AtMORC7 , we performed mRNA-Sequencing ( RNA-seq ) on leaves from individual plants of Col-0 , atmorc4-1 , atmorc7-1 , and atmorc4-1 atmorc7-1 backgrounds ( hereafter referred to as wild-type ( wt ) , atmorc4 , atmorc7 and atmorc4/7 , respectively ) . We found that AtMORC4 and AtMORC7 affect a highly overlapping gene set with AtMORC7 playing a more dominant role ( Fig 1C–1E ) . In atmorc7 , 348 annotated loci were differentially expressed ( FDR < 0 . 05 ) with 84% being up-regulated . In atmorc4 , the 33 differentially expressed loci ( 30 up , 3 down ) were largely a subset of those altered in atmorc7 , with 29 of the 30 up-regulated loci also up-regulated in atmorc7 . In the atmorc4/7 double knockout , 50% more loci were differentially expressed than in the individual knockouts combined , suggesting a significant level of redundancy between AtMORC4 and AtMORC7 . Taken together , the results suggest that AtMORC4 and AtMORC7 act in a partially redundant manner , with AtMORC7 having a stronger effect than AtMORC4 , to mainly repress a highly overlapping gene set . We have previously shown that AtMORC6 forms mutually exclusive heteromeric complexes with either AtMORC1 or AtMORC2 [35] . To assess whether AtMORC4 and AtMORC7 form heteromeric complexes , we generated endogenous promoter driven MYC or FLAG tagged lines for both AtMORC4 and AtMORC7 in their respective T-DNA backgrounds . By co-immunoprecipitation , we detected a homotypic association of AtMORC4 and AtMORC7 but did not detect heteromers ( Fig 2A–2C ) . These results were confirmed by mass spectrometry of the immunoprecipitated samples ( IP-MS ) , showing that the AtMORC4 and AtMORC7 precipitates do not contain peptides from AtMORCs other than themselves ( Fig 2D ) . Together , this indicates that AtMORC4 and AtMORC7 form homomeric complexes in vivo , consistent with the genetic redundancy observed between them ( see Fig 1 , S1 Fig ) . To directly compare the phenotypes of the atmorc4 and atmorc7 mutants with the previously characterized atmorc6-3 ( hereafter referred to as atmorc6 ) , we performed a second round of RNA-seq analysis . We also sought to generate a genetically MORC-less plant to obtain an unobfuscated view of MORC function . For this , we created a higher order knockout plant containing T-DNA inserts in six out of the seven MORC genes in Arabidopsis , atmorc1-2 , atmorc2-1 , atmorc4-1 , atmorc5-1 , atmorc6-3 , and atmorc7-1 ( atmorc1/2/4/5/6/7 ) . While a previous study reported embryonic lethality for a T-DNA insertion in AtMORC3 [NP_195350; AT4G36270; CRH2] [36] , it is likely that this is an indirect effect caused by an unknown linked mutation in the SALK line ( SALK_000009 ) , as we found evidence suggesting that AtMORC3 is in fact a pseudogene ( S2 Fig ) . We found a premature stop codon in exon three in Col-0 ( causing either an un-translated or truncated protein ) . Additionally , an independent homozygous T-DNA allele ( SALK_043244 ) with an exonic insertion exhibited no discernable phenotype . Given that AtMORC3 is non-functional in Col-0 , the atmorc1/2/4/5/6/7 line effectively lacks any functional AtMORC protein . RNA-seq on individual plants ( 3 replicates each ) from atmorc6 , atmorc4/7 , atmorc4/6/7 , and atmorc1/2/4/5/6/7 revealed 39 , 815 , 1188 , and 1519 differentially expressed genes ( FDR < 0 . 05 ) relative to wt , respectively , with a variety of interesting features ( Fig 3 ) . Twenty times more loci were differentially expressed in atmorc4/7 as compared to atmorc6 , suggesting that AtMORC4 and AtMORC7 play a more central role in gene expression ( Fig 3A ) . As the majority of these atmorc4/7 differentially expressed genes were up-regulated ( 87% ) , this is consistent with a repressive role and direct regulation at these targets . However , we cannot exclude the possibility of indirect effects . The difference between atmorc6 and atmorc4/7 is also clearly apparent from a heatmap over the union set of differentially expressed loci , which shows that atmorc6 is most similar to wt ( Fig 3B ) . In atmorc6 , transposable elements ( TEs ) constitute 29% ( 11 total ) of the differentially expressed loci while in atmorc4/7 , only 1% ( 9 total ) were misregulated , suggesting that AtMORC6 is preferentially involved in TE repression while AtMORC4 and AtMORC7 are primarily responsible for the repression of protein-coding genes . Comparing atmorc4/7 to atmorc6 revealed that while there was a generally positive correlation , many loci are specifically affected in either atmorc6 or atmorc4/7 ( Fig 3C ) . One example is ZF1 , which encodes a stimulus response zinc finger protein characteristic of the types of genes up-regulated in atmorc4/7 ( see below ) and is up-regulated only in atmorc4/7 . On the other hand , the gene SDC [37] was much more highly up-regulated in atmorc6 than it was in atmorc4/7 , consistent with the use of its promoter in the forward genetic screen that resulted in isolation of atmorc6 [25] . A similar plot comparing atmorc4/6/7 versus atmorc1/2/4/5/6/7 showed an extremely close correlation ( Fig 3D and see S3 Fig ) . This demonstrates that AtMORC1 , AtMORC2 , and AtMORC5 [NP_196817; At5G13130; CRH5] do not have a significant impact on the transcriptome , consistent with the previous report indicating that atmorc1/2 is equivalent to that of atmorc6 and that the expression of AtMORC5 is pollen specific [35] . We performed GO term analysis on the genes misregulated in atmorc4/7 , which revealed a striking enrichment for immune response genes , especially ‘response to chitin’ ( p value = 2 . 3e-47 ) ( S4 Fig ) . Interestingly , we had previously noted ‘response to chitin’ , albeit with lower significance , ( p < 6e-4 ) , for genes misregulated in atmorc6 [35] . Chitin is a component of the fungal cell wall and acts as a basal defense response elicitor [38] . In addition , AtMORC7 appears in an RNA co-expression network with multiple disease resistance genes , including LURP1[39] , PUB12[40] , ACD6[41] , SDE5[42] and three NB-LRR type proteins [43] ( Fig 4A ) . Since LURP1 mutants are compromised in defense against the Emwa1 isolate of the oomycete pathogen , Hyaloperonospora arabidopsidis ( Hpa ) [39] and atmorc1 was also identified as showing enhanced susceptibility to this pathogen [44] , we challenged atmorc1 , atmorc6 , atmorc4 , atmorc7 , atmorc4/7 , atmorc4/6/7 and atmorc1/2/4/5/6/7 with Emwa1 Hpa . We observed significantly increased susceptibility in atmorc1 , atmorc6 , atmorc4/7 , atmorc4/6/7 and atmorc1/2/4/5/6/7 as compared to wt ( Fig 4B ) . The individual atmorc4 and atmorc7 mutants did not show a difference from wild type , providing further support for the functional redundancy between AtMORC4 and AtMORC7 . As we did not observe an additive increase in susceptibility in the higher order atmorc mutants , we reasoned that this might reflect non-additive changes in the transcriptome . Indeed , the atmorc4/6/7 and atmorc1/2/4/5/6/7 plants showed no further increase in expression of the ‘response to chitin’ ( GO:0010200 ) gene set than did atmorc4/7 ( S5 Fig ) . While the mis-expression of specific genes in this set may contribute to pathogen susceptibility , it also remains possible that AtMORC proteins play a more direct role in defense [31 , 36 , 45] . Together , these results suggest that—in addition to AtMORC1—AtMORC6 , AtMORC4 , and AtMORC7 act as positive regulators of defense in A . thaliana against the oomycete Hpa . In Arabidopsis , interphase chromosomes are organized into distinct chromosomal territories , with euchromatic arms looping out from condensed heterochromatic chromocenters [46–48] . These chromocenters constitute repeat and transposon-rich pericentromeric heterochromatin and are readily visible by light microscopy as intensely DAPI stained nuclear foci . AtMORC1 and AtMORC6 form punctate bodies adjacent to chromocenters and in atmorc6 mutants , pericentromeric regions are decondensed , suggesting that AtMORC6 plays a role in higher order chromatin compaction at the interface of these transposon-rich regions [25 , 48] . Because AtMORC4 and AtMORC7 were found to target both genes and transposons , we determined their localization in the nucleus . Using pAtMORC4::AtMORC4-MYC and pAtMORC7::AtMORC7-MYC lines , we observed chromocenter adjacent bodies formed by both AtMORC4 and AtMORC7 ( Fig 5A and 5B and S1 and S2 Videos ) . AtMORC7 bodies were generally more intensely stained than AtMORC4 bodies . Consistent with the effects of atmorc4/7 mutation on euchromatic gene expression , AtMORC4 and AtMORC7 were also uniformly distributed throughout the nucleoplasm whereas AtMORC1 and AtMORC6 tended to appear as punctate nuclear foci ( see Fig 5C and 5D and previously observed [25] ) . AtMORC4 and AtMORC7 staining was specifically excluded from chromocenters , but was frequently enriched along chromocenter boundaries , forming multiple foci or forming rings around chromocenters ( Fig 5 ) . The function of these nuclear bodies is currently unknown . We utilized the atmorc1/2/4/5/6/7 hextuple mutant to determine the contribution of AtMORCs to DNA methylation patterning . We performed whole-genome bisulfite sequencing ( BS-seq ) , to examine DNA methylation at single cytosine resolution , in atmorc1/2/4/5/6/7 as well as atmorc4/7 and wt ( 2 biological replicates each ) . We also included the previously published BS-seq dataset for atmorc6 [35] in our analysis . Global levels of methylation over the chromosomes were unaltered in any AtMORC knockout background in all three sequence-contexts ( S6A Fig ) . Focusing specifically on loci that were de-repressed in atmorc1/2/4/5/6/7 , we observed very little overall change in methylation upstream , downstream or throughout the gene body at these loci ( S6B Fig ) . These results suggest that the most significant changes in transcription resulting from the loss of AtMORCs are not generally accompanied by losses in DNA methylation . Next we examined the potential contribution of AtMORC to the different DNA methylation pathways . MET1 maintains CG methylation throughout the genome , CMT3 maintains the majority of CHG methylation , DRM2 maintains CHH methylation at RdDM sites , and CMT2 maintains CHH methylation in pericentromeric heterochromatin [3 , 7 , 8 , 10] . Using previously defined loci whose methylation is dependent upon these methyltransferases [8 , 49] , we examined methylation levels in the AtMORC mutants . Again we found essentially no reduction in methylation in the AtMORC knockouts , suggesting that AtMORCs do not play a significant role in any of the major DNA methylation pathways in Arabidopsis ( S7A Fig ) . We also tested whether AtMORCs might act downstream of DNA methylation from any of these specific methyltransferase pathways by plotting RNA-seq reads over differentially methylated regions ( DMRs ) defined as changing in the different methyltransferase mutant backgrounds; however , we did not observe any consistent changes in bulk levels of RNA in the AtMORC knockouts at these collections of methylated loci ( S7B Fig ) . Since AtMORC6 has been implicated in transcriptional silencing at RdDM loci , reportedly interacting with members of the RdDM pathway [19 , 26] , we examined whether there might be more localized changes in DNA methylation by parsing the genome into 100bp windows and searching for DMRs . We found 519 atmorc1/2/4/5/6/7 hypomethylated CHH DMRs , 54% of which overlapped with drm1/2 hypomethylated CHH DMRs ( Fig 6A , S8A Fig ) . In addition , the remaining 46% of hypomethylated CHH DMRs that were called as being specific to atmorc1/2/4/5/6/7 in fact showed dramatically reduced methylation in drm1/2 ( Fig 6B , right panel ) , suggesting that even though these DMRs did not make the stringent cutoff required to be a DMR , the majority of atmorc1/2/4/5/6/7 hypomethylated DMRs correspond to sites of RNA directed DNA methylation . In contrast , only 2% of atmorc1/2/4/5/6/7 hypomethylated DMRs exclusively overlapped with cmt2 hypomethylated CHH DMRs ( S8A Fig ) . We also checked whether these atmorc1/2/4/5/6/7 hypomethylated CHH DMRs might be the result of spontaneous epi-allelic variation by comparison with a previously defined set of DMRs that are known to change states in the wild type [50] , but found only a 3% overlap ( S8B Fig ) . Together , these data suggest that AtMORCs are required for CHH methylation at a small subset of drm1/2-RdDM loci . Comparing atmorc6 with atmorc4/7 at atmorc1/2/4/5/6/7 hypo CHH DMRs , we found that atmorc6 more strongly resembles that of atmorc1/2/4/5/6/7 ( S9 Fig ) . Interestingly , atmorc4/7 and atmorc6 do not appear to affect mutually exclusive regions , suggesting that AtMORC4/7 and AtMORC6 are required at overlapping target loci ( S9A Fig ) . However , atmorc4/7 generally showed less severe CHH methylation loss than atmorc6 ( S9A and S9B Fig ) , which is consistent with AtMORC4 and AtMORC7 being primarily involved in repression of protein-coding genes , and AtMORC6 being predominantly involved in repression of methylated elements . Since the AtMORCs appear to be transcriptional repressors , we plotted RNA-seq data over the atmorc1/2/4/5/6/7 hypomethylated CHH DMRs . We observed a clear increase in bulk levels of RNA over these sites in the atmorc1/2/4/5/6/7 knockout ( S10A Fig ) . While this result might seem intuitive , this was not the case for drm1/2 hypomethylated CHH DMRs , where loss of DRM1/2 did not result in significant transcriptional re-activation ( S10B Fig and [8] ) . To determine whether the overall change in transcription seen in atmorc1/2/4/5/6/7 knockout is caused by a small number of jackpot sites or is the result of many DMRs becoming transcriptionally reactivated at a moderate level , we plotted RNA-seq reads from individual DMRs ( Fig 6C and 6D ) . We found that atmorc1/2/4/5/6/7 hypomethylated CHH DMRs were frequently characterized by transcriptional de-repression , while drm1/2 exclusive hypomethylated CHH sites were not . Interestingly , the atmorc1/2/4/5/6/7 defined hypomethylated CHH sites were also transcriptionally reactivated in the drm1/2 background ( Fig 6D ) . Thus this set of sites is susceptible to transcriptional depression when CHH methylation is lost , either by loss of RdDM or by loss of MORC function . In order to determine if the 519 atmorc1/2/4/5/6/7 hypomethylated DMR regions might have unique qualities that distinguish them from other sites that do not lose CHH methylation , we analyzed their DNA sequence composition . Interestingly , when we calculated CG , CHG , and CHH density , we found that the atmorc1/2/4/5/6/7 defined subset had significantly fewer CG and CHG sites as compared to the rest of the RdDM loci and compared to the genome average ( Fig 7 ) . An attractive hypothesis therefore is that a low density of symmetric methylation ( due to a low density of methylatable sites ) may not be sufficient to maintain silencing once asymmetric CHH methylation is lost , which would explain why these particular regions become reactivated in drm1/2 . Since AtMORCs are not generally required for CHH methylation maintenance , it would then seem likely that AtMORCs primary role would be to help maintain transcriptional repression at these regions of diffuse symmetric methylation and poised transcriptional potential . The transcriptional reactivation of these sites in atmorc may then secondarily lead to loss of CHH methylation at these loci , and it is indeed known that positive epigenetic marks associated with transcription can lead to a loss of RdDM function [14 , 51 , 52] . In addition , symmetric CG methylation plays a role in the stable association of Pol V to chromatin , and thus perpetuates RdDM and CHH methylation [18] . Thus we hypothesize that this unique set of 519 atmorc1/2/4/5/6/7 hypomethylated DMR regions experience a loss of methylation because they are both depleted in symmetric methylation and because they become transcriptionally reactivated in atmorc mutants . In this study , we established a role for the previously uncharacterized AtMORC4 and AtMORC7 genes in widespread repression of protein-coding genes and in pathogen defense . We found that these proteins act partially redundantly , forming mututally exclusive homomeric complexes , which explains why they have not previously been identified in forward genetic screens . In addition , AtMORC4 and AtMORC7 formed bodies adjacent to chromocenters while also showing localization throughout the nucleoplasm . By analysing a compound mutant devoid of all MORC function , we showed that AtMORC is not a key component in the maintenance of any of the major DNA methylation pathways and that major changes in transcription were not generally accompanied by loss of DNA methyation . However , at a small subset of RdDM targets ( approximately 5% ) , AtMORC was required for both methylation and silencing , suggesting that these methylation losses are likely an indirect consequence of the loss of gene silencing . These findings reconcile our laboratory’s previous reports of methylation-independent silencing [25] with that of other laboratories reporting hypomethylation at specific de-repressed reporter loci in atmorc6 mutant backgrounds [26 , 27] . We recently reported that mouse MORC1 is required for DNA methylation and silencing at a specific subset of transposon promoters that are normally methylated at a developmentally late stage during the wave of global de novo methylation in the male germ line [53] . As in Arabidopsis , there were no genome wide changes in DNA methylation in the mouse morc1 mutant , but specific methylation defects at a class of transposons that failed to establish silencing . These commonalities suggest that Arabidopsis MORCs may act similarly to mammalian MORC1 , to maintain silencing at loci that are poised for transcriptional de-repression , with DNA hypomethylation as a secondary effect . Nuclear localization of AtMORC4 and AtMORC7 broadly reflected that of their euchromatic gene and pericentromeric transposon targets , with both chromocenter adjacent enrichment and distribution throughout the nucleus . Since we previously reported that AtMORC6 and AtMORC1 form chromocenter adjacent bodies [25] ( and see Fig 5 ) , this appears to be a general feature of Arabidopsis MORC proteins , although the function of these bodies is at present completely unknown . In the future , it will be important to determine the precise molecular mechanisms by which MORC proteins interact with chromatin and regulate gene expression .
Wild-type and all mutant lines are from the ecotype Columbia ( Col-0 ) and were grown under either continuous light ( S1 Fig , Fig 2 ) or long days ( 16 hour light—all other experiment ) . The T-DNA lines used in this study were: atmorc1-2 ( gene AT4G36290 ) SAIL_893_B06 ( aka crt1-2 ) , atmorc2-1 ( gene AT4G36280 ) SALK_072774C ( aka crh1-1 ) , atmorc3-2 ( gene AT4G36270 ) SALK_043244 , atmorc4-1 ( gene AT5G50780 ) GK-249F08 ( aka crh4-2 ) , atmorc5-1 ( gene AT5G13130 ) SALK_049050C ( aka crh5-2 ) , atmorc6-3 ( gene AT1G19100 ) GABI_599B06 ( aka crh6-5 ) , and atmorc7-1 ( gene AT4G24970 ) SALK_051729 ( aka crh3-1 ) . T-DNAs were confirmed by PCR based genotyping . Primer sequences are described in S1 Table . The pAtMORC4::AtMORC4-MYC , pAtMORC4::AtMORC4-FLAG , pAtMORC7::AtMORC7-MYC , and pAtMORC7::AtMORC7-MYC constructs were generated by the same method described in [35] . Briefly , the AtMORC4 and AtMORC7 genomic regions , including ~1 kb upstream from the transcriptional start sites , were PCR amplified and cloned into a pENTR/D-TOPO vector ( #K2400-20 , Thermo Fisher ) . The cloned genomic regions were then transferred into a pEG302 based binary destination vector that included a MYC or FLAG epitope tag at the C-terminus via a Gateway LR Clonase II reaction ( #11791–100 , Thermo Fisher ) . Agrobacterium tumfaciens AGLO strain carrying these constructs were used to transform A . thaliana plants in their respective mutant backgrounds using the floral dip method [54] . 2–3 leaves from individual 3-week old plants were used to make individual BS-seq libraries based on methods described by [49] . Briefly , genomic DNA was extracted using DNeasy Plant Mini kit ( #69106 ) and 500ng was sheared using the Covaris S2 instrument . Libraries were generated using the Kapa Hyper Prep Kit ( #KK8502 ) with bisulfite conversion using the EZ DNA Methylation Lightning Kit ( #D5030 ) . Libraries were sequenced on a HiSeq 2000 ( Illumina ) . RNA was extracted from 2–3 leaves of 3-week old plants using Trizol reagent and DNAse treated using TURBO DNA-free kit ( #AM1907 ) . For RNA-seq , 1–2 . 5 μg of RNA starting material per library was first rRNA depleted using Epicentre RiboZero ( #MRZPL1224 ) prior to library generation using Epicentre ScriptSeqv2 ( #SSV21124 ) . Libraries were sequenced on a HiSeq 2000 ( Illumina ) . For RT-PCRs , cDNA was generated using SuperScript III ( #18080–044 , ThermoFisher ) with random hexamer priming . The samples were digested with RNAse H in accordance with manufacturer’s protocol . RT–PCR was then performed with iQ SYBR Green Mastermix ( BioRad ) using an Agilent Technologies Mx3005p qPCR System ( Stratagene ) . Hyaloperonospora arabidopsidis ( Hpa ) isolate Emwa1 was propagated on the susceptible Arabidopsis ecotype Ws . Conidiospores of Hpa strain Emwa1 were resuspended in autoclaved RO-water at a concentration of 3×104 spores/mL and spray-inoculated onto 10-day old seedlings . Inoculated plants were covered with a lid to increase humidity and grown at 19°C under a 9-hour light period . Sporangiophores per cotyledon were counted 4 to 5 days post inoculation using a Leica M205 FA stereoscope . The experiments were repeated 3 times and the sporangiophores on approximately 100 cotyledons per genotype were counted in each experiment . Co-IP and IP-MS on pAtMORC4::AtMORC4-MYC/FLAG and pAtMORC7::AtMORC7-MYC/FLAG lines were performed as previously described [35] . For IP-MS , M2 magnetic FLAG-beads ( SIGMA , M8823 ) were added to the supernatant and immunoprecipitated proteins were eluted using 3×FLAG peptides ( SIGMA , F4799 ) . The MS was performed as described by [55] . For the Co-IPs , we added 100 μL M2 magnetic FLAG-beads ( SIGMA , M8823 ) to the supernatant for pulldown . For the western blots , we used HRP-coupled FLAG-specific antibody ( SIGMA , A8592 ) and MYC-specific antibodies ( Pierce , MA1-980 ) . Nuclear immunofluorescence experiments for AtMORC4/7-MYC tagged lines were performed based on the method described in [25] . Leaves from three-week old plants were fixed in 4% paraformaldehyde in TRIS buffer ( 10 mM TRIS pH 7 . 5 , 10 mM EDTA , and 100 mM NaCl ) for 20 minutes and washed twice in TRIS buffer . Leaves were chopped in 200–400 microliters lysis buffer ( 15 mM TRIS pH 7 . 5 , 2 mM EDTA , 0 . 5 mM spermine , 80 mM KCl , 20 mM NaCl , and 0 . 1% Triton X-100 ) and filtered through a 3 μM cell strainer ( Corning , #352235 ) . 5 μL of nuclei suspension was added to 12 μL of sorting buffer ( 100mM TRIS pH 7 . 5 , 50mM KCl , 2mM MgCl2 , 0 . 05% Tween-20 , and 20 . 5% sucrose ) and air dried on chloroform dipped microscope slides for two hours and then post-fixed in 4% paraformaldehyde in PBS for 20 minutes . Slides were washed three times in PBS and incubated in blocking buffer ( 3% BSA , and 10% horse serum in PBS ) for 30 minutes at 37°C . Nuclei were incubated at 4°C overnight in mouse monoclonal antibody against c-Myc ( 9E10 , Abcam ab32; 1:200 ) . Slides were washed in PBS and incubated with goat anti-mouse FITC antibody ( Abcam , ab7064; 1:200 ) for 90 minutes at room temperature . Following PBS washes , nuclei were counterstained and mounted in Vectashield mounting media with DAPI ( Vector , H-1200 ) . Nuclei were analyzed with a Zeiss LSM 710 Confocal microscope at 63X or 100X magnification using Zen software . For RNA-seq analysis , reads were aligned with TopHat , including the fr-secondstrand parameter . Cufflinks was used to generate count data using annotation from TAIR10 that was fed into the DEseq2 package in R for differential expression analysis . For BS-seq , reads were aligned using BSMAP with methylation levels calculated and DMRs defined as previously described [49] . For the atmorc DMRs , each biological replicated ( two per mutant ) was compared against two wild type biological replicates from the same experiment , requiring that the DMR be identified in all four mutant vs . wt comparisons to be considered a ‘true’ DMR . The dmr1/2 , cmt2 , cmt3 , and met1 DMRs were previously defined [49] , using a single mutant biological replicate compared against three biological wild type replicates . The data reported in this paper have been deposited in the Gene Expression Omnibus ( GEO ) database ( accession number GSE78836 ) . | Keeping selfish genetic elements–such as transposons–silent , while maintaining access to genes , is a fundamental challenge for eukaryotes . Different pathways frequently converge in order to identify transposons and maintain their repression , and in Arabidopsis thaliana , transposons are marked with DNA methylation . Previous studies of the Arabidopsis MORC proteins , which represent a highly conserved protein family , showed that AtMORC1 , AtMORC2 , and AtMORC6 are required for repression of methylated target transposons . Here , we describe the Arabidopsis genes AtMORC4 and AtMORC7 , which , instead of targeting methylated elements , appear to act redundantly to repress a large set of protein-coding genes and are required to mount a full defense against pathogen challenge . These proteins localize throughout the nucleus and form punctate bodies at the boundaries of highly compacted chromatin . By knocking out all functional copies of MORC genes in Arabidopsis , we find that major changes in transcription are not generally associated with the loss of DNA methylation . However , MORC may be recruited to assist in silencing of methylated regions that are unusually susceptible to transcriptional re-activation . This indicates that MORC and DNA methylation are convergently required to maintain repression at transposon targets . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"nuclear",
"staining",
"brassica",
"dna",
"transcription",
"model",
"organisms",
"genetic",
"elements",
"epigenetics",
"dna",
"plants",
"dna",
"methylation",
"chromatin",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"chromosome",
"biology",
"gene",
"expression",
"chromatin",
"modification",
"dna",
"modification",
"genetic",
"loci",
"biochemistry",
"dapi",
"staining",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"genomics",
"mobile",
"genetic",
"elements",
"organisms"
] | 2016 | Arabidopsis AtMORC4 and AtMORC7 Form Nuclear Bodies and Repress a Large Number of Protein-Coding Genes |
The expression of protein-coding genes requires the selective role of many transcription factors , whose coordinated actions remain poorly understood . To further grasp the molecular mechanisms that govern transcription , we focused our attention on the general transcription factor TFIIH , which gives rise , once mutated , to Trichothiodystrophy ( TTD ) , a rare autosomal premature-ageing disease causing inter alia , metabolic dysfunctions . Since this syndrome could be connected to transcriptional defects , we investigated the ability of a TTD mouse model to cope with food deprivation , knowing that energy homeostasis during fasting involves an accurate regulation of the gluconeogenic genes in the liver . Abnormal amounts of gluconeogenic enzymes were thus observed in TTD hepatic parenchyma , which was related to the dysregulation of the corresponding genes . Strikingly , such gene expression defects resulted from the inability of PGC1-α to fulfill its role of coactivator . Indeed , extensive molecular analyses unveiled that wild-type TFIIH cooperated in an ATP-dependent manner with PGC1-α as well as with the deacetylase SIRT1 , thereby contributing to the PGC1-α deacetylation by SIRT1 . Such dynamic partnership was , however , impaired when TFIIH was mutated , having as a consequence the disruption of PGC1-α recruitment to the promoter of target genes . Therefore , besides a better understanding of the etiology of TFIIH-related disease , our results shed light on the synergistic relationship that exist between different types of transcription factors , which is necessary to properly regulate the expression of protein coding genes .
In response to various physiological signals , selective and combined actions of a number of transcription factors modulate the expression of protein-coding genes [1] . In eukaryotes , the combinatorial use of hundreds of proteins is required for the synthesis of a single messenger RNA by the RNA Polymerase II ( RNA Pol II ) in association with the general transcription factors TFIIA , B , D , E , F and H [2] . Such number of protagonists requires dynamic networks to coordinate their actions during transcription . This is notably the case when a critical physiological parameter as glycaemia must be preserved within a narrow range . Indeed , to avoid the deleterious effects of hypo or hyperglycemia , the organism maintains steady circulating glucose levels , providing glucose for cells dependent on this fuel , such as neuronal and red blood cells . Apart from modulation of enzymes activity through posttranslational modifications and allosteric controls , some transcriptional regulations of rate limiting enzymes are intimately involved in maintaining blood glucose levels . Such transcriptional control is fundamental in the liver , which plays a central role in integrating signals of several cell types and multiple metabolic pathways . In particular , during starvation , in response to physiological signals like glucagon and glucocorticoids , hepatic gluconeogenic genes ( such as the phosphoenolpyruvate carboxykinase Pepck and the glucose-6-phosphatase G6Pase ) are regulated through the selective action of many transcription factors [3] , including cAMP-responsive element binding protein ( CREB ) [4] in association with the CREB-regulated transcription coactivator 2 ( CRTC2/TORC2 ) [5] , CCAAT enhancer-binding protein ( C/EBP ) [6] , Forkhead O box 1 ( FOXO1 ) [7] , [8] , hepatocyte nuclear factors ( especially HNF-4α ) [9] and the glucocorticoid receptor ( GR ) [10] , [11] . The transactivation mediated by these transcription factors is potentiated synergistically by different coactivators , such as the peroxisome proliferator-activated receptor-α coactivator 1 ( PGC-1α ) . Interestingly , although its role on gluconeogenic gene remains unclear [12] , [13] , PGC-1α gene expression during fasting is rapidly induced in mouse liver , concomitantly with PEPCK and G6Pase up-regulation [14] , [15] . Strikingly , PGC-1α activity is closely dependent to its deacetylation by the energy sensor SIRT1 ( silent mating type information regulation 2 homolog 1 ) , which is stimulated by the intracellular NAD+ and indirectly by pyruvate high concentrations related to starvation [15] , [16] . Many transcription factors have been identified during the fasting-induced expression of gluconeogenic genes , but little is known about their cooperative interactions with the general transcription machinery . Among the basal transcription factors , TFIIH plays a pivotal role during transcription by interacting with different factors including nuclear receptors [17] . TFIIH , which is also intimately implicated in the nucleotide excision repair ( NER ) pathway , is a multienzymatic protein complex that can be resolved into two subcomplexes: the core ( containing the helicase XPB , p62 , p52 , p44 , p34 and TTDA ) and the cdk-activating kinase complex ( CAK , containing MAT1 , CYCLIN H and the cyclin-dependent kinase CDK7 ) . While the helicase XPD subunit , which bridges the CAK to the core of TFIIH , mainly functions in NER [18] , the XPB helicase catalyzes DNA unwinding around the transcription initiation site and contributes to the promoter escape by RNA pol II [19] . In addition , TFIIH owns a kinase activity ( via its CDK7 subunit ) that contributes to transcription initiation by phosphorylating the C-terminal domain ( CTD ) of the RNA pol II largest subunit [20] . The phosphorylation by CDK7 is also required for the optimal transactivation mediated by different nuclear receptors , such as the peroxisome proliferator activated receptors PPARs [21] and the thyroid hormone receptors TR [22] . The key role played by TFIIH is illustrated by the fact that several mutations in its XPB , XPD and TTDA subunits lead to the rare autosomal recessive disorders Xeroderma Pigmentosum ( XP ) , sometimes associated with Cockayne syndrome ( XP/CS ) , and Trichothiodystrophy ( TTD ) . Besides photosensitivity and skin cancers , these patients can exhibit a large spectrum of clinical abnormalities , including skeletal defects , mental retardation , immature sexual development and dwarfism . To gain more insight into these complex clinical symptoms , mouse models bearing TFIIH mutations found in patients have been generated . In particular , the TTD mouse model ( having the most common XPD/R722W point mutation found in TTD patients ) [23] develops a phenotype similar to what observed in patients [24] including , beside the typical dry and brittle hairs [25] , progressive cachexia and hypoplasia of the adipose tissues [26] . TTD mice weight loss is not related to aberrant food uptake or intestinal malabsorption [21] , and analysis of blood cell parameters of adult mice only revealed a mild normochromic anaemia and decreased levels of branched-chain amino acids that were potentially related to starvation [24] . TTD mice thus suffer from progressive failure to thrive , which is likely to be the cause of premature death , but the major cause of this failure is unknown . Although TTD has been defined at first as a DNA repair syndrome [27] , transcriptional impairments may have a role in the development of TTD phenotypes [17] . In this regard , it has been shown that PPARγ dysfunction contributes to the hypoplasia of the adipose tissues [21] . However , the pleiotropic nature of the TTD phenotypes might imply combinational defects of other additional transcription factors . To study the transcriptional defects that might occur in TFIIH related diseases , TTD mice were subjected to different fasting periods . According to the fact that the physiological adaptation to fasting requires accurate genes regulation in the liver , the hepatic response to starvation of TTD mice was particularly studied . Taken together , our results show that TTD mutation disrupts the dynamic partnership between TFIIH , SIRT1 and PGC-1α , thus impeding the PGC-1α deacetylation by SIRT1 and consequently the correct expression of gluconeogenic genes .
While TTD mice suffer from progressive failure to thrive , the young adults mice did not yet develop severe metabolic alterations when normally fed [21] , [24] ( ref therein ) . Having observed that the daily food intake was identical between 3-months-old WT and TTD mice ( Figure 1A ) , the two lines have been then subjected to different periods of fasting ( 24 and 48 hours ) . While the body weight was progressively decreased in both strains ( Figure 1B ) , the liver weight was reduced in WT and TTD mice after 24 hours ( reduction of 14 . 4%±6 . 5% in WT and 17 . 3%±3 . 6% in TTD , respectively ) , followed by a higher reduction in TTD mice , losing almost 40% ( 39 . 8%±9 . 2% ) of the weight after prolonged fasting ( 48 hours ) ( Figure 1C ) . Different serological analyses were performed and revealed slight differences after starvation between the two strains , especially for triglycerides , free fatty acids and ketones bodies ( as illustrated by β-hydroxybutyrate ) ( Figure 1E , F , H ) , which might be related to the lower fat mass already observed in TTD fed normally ( Figure 1D ) [21] . In parallel , fasting blood lactate levels were similar between the two lines ( Figure 1I ) [28] , suggesting that lactate was normally utilized in TTD mice as a major substrate for glucose synthesis via gluconeogenesis . Finally , insulin and glucose , at a comparable level at fed state in WT and TTD mice , were similarly reduced during fasting ( Figure 1K and 1L ) , while glucagon progressively increased in WT and TTD ( Figure 1J ) . In parallel , to evaluate the gluconeogenic potential of TTD mice , pyruvate tolerance tests were undertaken ( Figure 1M ) . After injection of the gluconeogenic substrate pyruvate , TTD mice showed significant lower plasma glucose levels at 30 , 45 , 60 and 80 min , after what they reached the same levels of that observed in WT mice . Taken together , these data suggest that , although able to maintain physiological glucose level , TTD mice might develop subtle gluconeogenic defects upon fasting . Since the liver plays a central role in gluconeogenesis , we next focused our investigation on the hepatic response to fasting in TTD mice . We previously observed no significant difference in the architecture of the hepatic parenchyma ( as revealed after H&E staining , Figure 1G , sections 1 and 2 ) between WT and TTD mice fed normally [21] . However , glycogen content , visualized through Periodic Acid-Schiff staining ( sections 3–8 ) , showed slight preferential accumulation in pericentral hepatocytes ( located around pericentral veins , CV ) of WT mice fed normally ( section 3 ) , while such glycogen accumulation seemed to be more pronounced in TTD livers ( compare sections 3 and 4 ) . As expected , the glycogen stores were totally depleted in WT and TTD livers after 24 h of fasting ( sections 5 and 6 ) . Afterwards , glycogen was re-accumulated in periportal hepatocytes ( located around periportal veins , PV ) during a prolonged fasting period ( 48 h , sections 7 and 8 ) , indicating that glucose-6-phosphate was channeled towards glycogen , as previously observed in rodent livers [28] , [29] . However , such de novo glycogen accumulation was disorganized in TTD livers , with a higher number of hepatocytes bearing strong intracellular glycogen deposits when compared to WT ( compare sections 7 and 8 ) . Knowing that PEPCK and G6Pase are two key hepatic gluconeogenic enzymes required to meet energy demands during stressful conditions like starvation [3] , we analyzed their zonal distribution in TTD liver . Surprisingly , in normal feeding conditions immunohistochemical ( IHC ) staining of PEPCK showed a higher signal in hepatocytes located around portal vein ( PV ) in TTD liver ( Figure 2A , sections 1–2 ) . Prolonged fasting increased the PEPCK protein levels within the hepatic parenchyma , with a persistent higher signal in TTD when compared to WT ( sections 3–4 ) . In parallel , IHC staining of G6Pase revealed a low signal around central vein ( CV ) in livers of WT and TTD mice fed normally ( Figure 2B , sections 1 and 2 ) . After 48 h of fasting G6Pase protein level raised throughout the liver parenchyma of WT mice ( section 3 ) , whereas it did not increase in TTD ( section 4 ) . Knowing that the Pepck and G6pase genes are tightly regulated at a transcriptional level [14] , we analyzed the amount of their corresponding mRNA by quantitative RT-PCR . We repeatedly observed a higher amount of Pepck mRNA in liver of TTD mice fed ad libitum when compared to WT ( Figure 2C ) , which mirrored the protein amount observed by IHC analyses ( Figure 2A , sections 1–2 ) ; after fasting , the Pepck mRNA amount progressively increased in WT liver , whereas its accumulation occurred tardily in TTD liver , which might be related to the fact that basal levels at the feeding state were higher ( Figure 2C ) . In parallel , the amount of the G6pase mRNA progressively increased in the liver of WT , while its accumulation seemed to be delayed in TTD liver ( Figure 2D ) . These data prompted us to study the PGC-1α coactivator , which contributes to the hepatic transcriptional activation of the Pepck and G6pase genes during starvation [14] , [15] . Whereas Western Blot analyses showed an increase of the PGC-1α protein level in WT and TTD liver after 48 h of fasting ( Figure 2E , compare lanes 7–9 to lanes 1–3 and lanes 10–12 to lanes 4–6 , respectively ) , quantitative RT-PCR analyses revealed that the mRNA amount of the fasting-induced Pgc-1α gene was delayed in TTD liver when compared of that observed in WT ( Figure 2F ) . Taken together , these data suggest that defective expression of gluconeogenic genes occurs during fasting in TTD liver . To accurately dissect the transcriptional response occurring in TTD liver during fasting , hepatocytes were immortalized from WT and TTD mouse embryonic livers and then treated with medium devoid of glucose and supplemented with pyruvate , forskolin and glucagon , a treatment ( referred for convenience pyruvate treatment ) known to stimulate the gluconeogenic pathway [15] . Quantitative RT-PCR showed that the Pgc-1α gene was weakly induced 2 hours post-treatment in TTD hepatocytes when compared of that observed in WT hepatocytes ( Figure 3A1 ) . The expression of the PGC-1α-dependent Pepck and G6pase genes was lower and delayed in TTD cells when compared to their induction in WT hepatocytes ( Figure 3B1 and C1 , respectively ) . Remarkably , transfection of XPDwt in TTD hepatocytes restored the expression profiles of the Pgc-1α , Pepck and G6Pase genes similarly to that observed in normal cells ( Figure 3A1 , B1 and C1 ) , suggesting that the mutation in the xpd gene promoted the defective expression of gluconeogenic genes observed in TTD hepatocytes . It is worthwhile to notice that the expression of PGC1-α target genes involved in fatty acid oxidation , such as Cpt1a ( Carnitine palmitoyltransferase Ia ) and Mcad ( Medium-chain acyl-CoA dehydrogenase ) was also disrupted in TTD hepatocytes ( Figure S1A ) . Chromatin Immunoprecipitation ( ChIP ) assays then showed that RNA pol II and TFIIH ( visualized by the presence of its p62 and CDK7 subunits ) were recruited 2 hours post-treatment at the PGC-1α promoter ( Figure 3 , panels A2 to A4 ) and 4 h post-treatment at the PEPCK and G6Pase promoters ( panels B2–B4 and C2–C4 , respectively ) in WT hepatocytes , matching the expression profile of the corresponding genes . On the contrary , RNA pol II and TFIIH recruitments were altered and delayed in TTD hepatocytes . Interestingly , we also noticed that the recruitment of PGC-1α was disrupted on its own promoter as well as on that of PEPCK and G6Pase ( panels A5 , B5 and C5 ) . As a consequence , the recruitment profiles of transcription factors that are activated by PGC-1α , such as HNF-4α [9] , FOXO1 [30] and CREB [31] , were altered in TTD hepatocytes ( Figure S1B ) . It is worthwhile to notice that the protein amounts of HNF-4α , FOXO1 and CREB were similar in WT and TTD hepatocytes ( Figure S1C ) and the nuclear localization as well as the acetylation status of FOXO1 was not disrupted in TTD hepatocytes ( Figure S1D ) . Finally , the recruitment of the deacetylase SIRT1 , which activates PGC-1α [15] , was also disrupted in TTD hepatocytes ( Figure 3 , panels A6 , B6 and C6 ) . Strikingly , Western Blots analyses revealed that the PGC-1α and SIRT1 protein amounts were similar in WT and TTD hepatocytes ( Figure 3D ) , suggesting that their defective recruitments on the promoter of target genes in TTD were not due to lower protein amounts . Nonetheless , we also noticed that the amount of TFIIH subunits ( such as p62 and CDK7 ) was reduced in TTD cells ( Figure 3D ) [32] . Taken together , these data show that the expression of PGC-1α is altered in TTD hepatocytes , resulting in a defective regulation of PGC-1α-targeted gluconeogenic genes . As a consequence , intracellular glucose 6-phosphate ( Figure 3E ) and glucose output ( Figure 3F ) were reduced in TTD hepatocytes after 12 h of pyruvate treatment . Since TFIIH and PGC-1α have in common the ability to coactivate some nuclear receptors [22] , [33] , [34] , and the activation by the latter requires its deacetylation by SIRT1 [15] , [16] , we investigated the partnerships that might exist between TFIIH , SIRT1 and PGC-1α . Having observed that TFIIH mutation disrupted PGC-1α and SIRT1 recruitment at the promoter of target genes ( Figure 3A5–6 , B5–6 and C5–6 ) , we investigated whether these factors interacted with each other . Antibodies directed towards PGC-1α co-immunoprecipitated SIRT1 as well as TFIIH ( illustrated by the presence of its p62 subunit ) from WT hepatocytes nuclear extracts ( Figure 4A , lane 2 ) . Strikingly , these co-immunoprecipitations were reduced in TTD hepatocytes extracts ( compare lanes 2 and 4 in Figure 4A ) . However , supplementation of TTD nuclear extracts with highly purified recombinant TFIIH ( rIIH ) potentiated the binding of SIRT1 to PGC-1α to a level similar to that observed in WT ( Figure 4A , lanes 4 and 5 ) . In a second set of assays , we observed that rIIH co-immunoprecipitated with purified PGC-1α ( Figure 4B ) [35] and SIRT1 ( Figure 4C ) . PGC-1α specifically interacted with recombinant XPB , p34 and MAT1 subunits of TFIIH ( Figure 4D ) , while SIRT1 co-immunoprecipitated with XPB , p62 , cdk7 and MAT1 ( Figure 4E ) . PGC-1α acetylation profile was then analyzed in WT and TTD hepatocytes after pyruvate treatment . Western Blots revealed that the immunoprecipitated PGC-1α was deacetylated until 6 h post-treatment in WT hepatocytes ( Figure 4F , lanes 1–3 ) concomitantly to a higher co-immunoprecipitation of SIRT1 ( Figure 4G , lanes 1–3 ) , thereby allowing the recruitment of PGC-1α on targeted promoters ( Figure 3A5 , B5 and C5 ) . On the contrary , in TTD hepatocytes PGC-1α acetylation remained relatively high ( Figure 4F , lanes 5–6 ) , which was in accordance with the lower co-immunoprecipitation of SIRT1 ( Figure 4G , lanes 5–6 ) and with the weaker recruitment of PGC-1α on targeted promoters ( Figure 3A5 , B5 and C5 ) . Taken together these results show that PGC-1α and SIRT1 both interact with TFIIH , and suggest that the reduced amount of TFIIH in TTD hepatocytes affects the partnership between SIRT1 and PGC-1α and consequently PGC-1α deacetylation . The connection between TFIIH , PGC-1α and SIRT1 was further investigated . We first evaluated whether the acetylation status of PGC-1α influenced its interaction with TFIIH and SIRT1 ( Figure 5A ) . Non acetylated and acetylated PGC-1α were immunoprecipitated from WT hepatocytes and were incubated with SIRT1 and recombinant TFIIH ( rIIH ) . Our results revealed that SIRT1 as well as TFIIH ( illustrated by the presence of its p62 subunit ) were able to co-immunoprecipitate with PGC-1α regardless of its acetylation status ( lanes 2 and 4 ) . Furthermore , addition of a specific SIRT1 inhibitor ( EX-527 , 10 µM ) [36] , [37] did not affect the TFIIH/PGC-1α/SIRT1 complex , whichever the PGC-1α acetylation status ( lanes 3 and 5 ) . Having observed that TFIIH ( via its CDK7 kinase subunit ) phosphorylated in vitro SIRT1 and not PGC-1α ( Figure 5B ) , recombinant TFIIH containing wild-type XPD subunit ( rIIH-XPD/WT ) was immunoprecipitated ( with antibodies directed against the XPB subunit ) and incubated with SIRT1 and/or PGC-1α in the presence or absence of 100 nM ATP ( Figure 5C ) . We firstly noted that the interaction between SIRT1 and rIIH XPD/WT was not modified upon addition of either ATP ( lane 4 ) or a CDK7 kinase inhibitor ( CDK7 inh . , lane 5 ) , suggesting that the CDK7-mediated phosphorylation of SIRT1 did not influence its binding to TFIIH . On the contrary , addition of ATP disrupted the interaction of PGC-1α to rIIH XPD/WT ( lanes 6–7 ) . While the presence of PGC-1α potentiated the binding of SIRT1 to the immunoprecipitated rIIH XPD/WT ( lane 8 ) , we observed that addition of ATP promoted the release of PGC-1α and therefore of SIRT1 ( lane 9 ) . Such effect was not related to the CDK7 kinase activity , since the addition of CDK7 inhibitor did not affect the release of SIRT1 as well as of PGC-1α ( lane 10 ) . Knowing that the TTD mutated form XPD/R722W affects the integrity of TFIIH by weakening the binding of the CAK subcomplex to the core of TFIIH [18] , [38] , we next evaluated the consequences of such mutation on the TFIIH/SIRT1/PGC-1α complex formation ( Figure 5D ) . Whereas the binding of SIRT1 alone to the immunoprecipitated rIIH XPD/R722W was not modified even in presence of ATP ( lane 3–5 ) , we observed that ATP modulated the binding of PGC-1α ( lanes 7 and 9 ) . Furthermore , contrary to that observed with rIIH XPD/WT , the binding of SIRT1 to rIIH XPD/R722W was not enhanced by the presence of PGC-1α ( lane 8 ) , suggesting that the integrity of TFIIH is crucial for the optimal binding of SIRT1 . The core of TFIIH without XPD and the CAK subcomplex ( core-IIH no XPD ) was then immunoprecipitated using antibodies directed against XPB , and incubated with SIRT1 and/or PGC-1α ( Figure 5E ) . Although SIRT1 and PGC-1α bound the core-IIH ( lanes 3 and 6 , respectively ) , the binding of PGC-1α in the presence of SIRT1 was reduced when compared to that observed with rIIH XPD/WT ( Figure 5C , lane 8 ) , suggesting that XPD and the CAK subcomplex are required for accurate binding of PGC-1α . In parallel , the binding of SIRT1 to the core-IIH was not strongly potentiated by the presence of PGC-1α ( lane 8 ) . Interestingly , while ATP promoted the release of PGC-1α from the immunoprecipitated complex ( lane 7 ) , it did not modify the binding of SIRT1 ( lane 4 ) , suggesting that the ATP-dependent release of PGC-1α did not involve the CAK subcomplex . Furthermore , non-hydrolysable ATP analog ( ATP-γS , 100 nM ) promoted the release of PGC-1α ( and therefore of SIRT1 ) from the core-TFIIH ( lane 10 ) . Having noticed that both PGC-1α and SIRT1 interacted with the XPB ATP-binding subunit of TFIIH ( Figure 4D and 4E , respectively ) , we incubated these three proteins in the presence of ATP ( Figure 5F ) . After immunoprecipitation of XPB , we observed that ATP promoted the release of PGC-1α ( lanes 6 ) without affecting the binding of SIRT1 ( lane 4 ) . Our data prompted us to immunoprecipitate a mutated form of XPB that does not bind ATP ( XPB/K346R , Figure 5G ) [39] , [40] , [41] . While SIRT1 and PGC-1α interacted with this mutated form of XPB ( lanes 3 and 5 , respectively ) , no release of PGC-1α was observed in the presence of ATP ( lane 6 ) . Taken together , these results suggest that the binding of ATP to XPB influences the PGC-1α release without affecting the interaction with SIRT1 .
The production of glucose during fasting requires an extensive use of metabolites , including lactate , amino acids , triglycerides and ketone bodies . Although serological analyses from fasted TTD mice revealed slight reductions when compared to WT for triglycerides , free fatty acids and ketones bodies ( Figure 1E , F , H ) [21] , TTD mice seemed to be able to cope with prolonged fasting periods by maintaining glucose at a level similar to that observed in normal mice ( Figure 1L ) . However , such maintenance in front of a stressful condition resulting from food deprivation was accompanied by failures in the liver . In particular , disruption was quickly observed during pyruvate tolerance tests ( Figure 1M ) , which implies defects in the liver and certainly in other peripheral tissues , such as the adipose tissues . Defect in different metabolic processes might be implicated , such as lipid uptake , mitochondrial function and gluconeogenic pathway . Furthermore , the de novo hepatic accumulation of glycogen occurring after longer fasting period was also disrupted in TTD ( Figure 1G ) . Although the gluconeogenic enzymes defects observed in TTD livers ( Figure 2 ) might contribute to the glycogen accumulation abnormalities , further investigations should be performed in order to claim such assertion . To maintain blood glucose levels in conditions of food deprivation , a precise regulation of the hepatic genes that control glucose production is required . However , the accurate and timely expression of the gluconeogenic genes Pepck and G6pase was clearly disrupted in TTD liver ( Figure 2C , D ) , which can therefore explain the abnormal amount of PEPCK and G6Pase observed after fasting ( Figure 2A , B ) . Although the phenotypes observed in TTD might result from pleiotropic effects due to the dysfunctions of various transcription factors , the deregulation of the gluconeogenic genes seems at least in part to concern PGC-1α , a coactivator whose induction is also delayed in fasted TTD ( Figure 2F ) , and this raises the question on the role of the mutated TFIIH in such defect . By weakening the overall structure of TFIIH and its cellular concentration ( Figure 3D ) [18] , [38] , it seems that the XPD/R722W mutation prevents the capacity of TFIIH to correctly interact with PGC-1α and SIRT1 ( Figure 4A and 5C ) and to promote the PGC-1α deacetylation by SIRT1 ( Figure 4F ) . Thus , PGC-1α recruitment is impaired on the promoter of PEPCK and G6Pase ( Figure 3 B5 and C5 , respectively ) , having as a consequence the alteration of the concomitant recruitment of transcription factors such as HNF-4α , FOXO1 and CREB ( Figure S1B ) . Besides the fact that the integrity of TFIIH is essential for the PGC-1α coactivation function , the partnership of TFIIH with SIRT1 and PGC-1α also requires an ATP-dependent dynamic process via the XPB subunit ( Figure 5F ) . Interestingly , the helicase XPB subunit is known to harbor conformational change in the presence of DNA , which is facilitated by ATP hydrolysis , leading to a closed and stable XPB/DNA complex [41] , [42] . It seems increasingly clear that conformational changes in XPB resulting from the binding and/or the hydrolysis of ATP would allow new connections with DNA and proteins . Taken together , our observations highlight the fine and dynamic relationships that exist between the basal transcription machinery , the transcription factors and their cofactors , which allow the expression of protein coding genes at the right time and in the right amount . Previous studies showed that TFIIH participates to the transactivation mediated by different nuclear receptors by phosphorylating them [21] , [22] , [43] . Interestingly , the CDK7 kinase subunit of TFIIH also phosphorylates SIRT1 but not PGC-1α , at least in vitro ( Figure 4A ) . Whereas the analysis of truncated forms of SIRT1 suggested that the catalytic and the C-terminal part of SIRT1 might be phosphorylated by the CDK7 kinase , we failed to identify the targeted residues . Interestingly , different phosphorylated sites have been already found in SIRT1 [44] , some of them promoting the deacetylation activity of SIRT1 [45] . Although the role of the SIRT1 phosphorylation by TFIIH is far from being established ( Figure 6 ) , the finding of a dynamic connection between TFIIH and SIRT1 remains particularly relevant , since SIRT1 plays a critical role in health maintenance and metabolic stress responses [46] , [47] . In particular , it has been shown that , while the accumulation of DNA damage caused by oxidative stress contributes to human tissue ageing , strong evidence supports a role for SIRT1 in oxidative stress response by deacetylating transcription factors that regulate the expression of stress response genes [48] , [49] . Furthermore , recent data suggest that the transcriptional arrest following UV irradiation in cells bearing XPD mutations associated to the combined XP/CS syndrome results from an active and persistent heterochromatinization process mediated by SIRT1 [50] . Therefore , in addition to further understand the synergistic action of factors during gene expression , our results allow us to better apprehend the etiology of human diseases during which transcriptional mechanisms may be impaired .
This study was performed with the agreement of the French Ministry of Higher Education and Research ( permit number n°6754 ) . Furthermore , our study has been followed by the Ethical committee of the Institute , which is registered by the French National Committee for Ethics in Animal Experimentation n°17 . This study has been realized in the accredited animal house of the Institute , in compliance with European guidelines . Euthanasia using carbon dioxide has been realized on adult mice . Every effort was made to minimize suffering . The generation of the TTD mouse line XPD/R722W has been previously described [24] . Three-month-old WT and TTD males used in our experiments share identical genetic background ( 100% C57BL/6 ) and are littermates . Mice were either fed a standard chow with 5% ( w/w ) fat content ( R03 breeding diet , UAR , Villemoisson , France ) or fasted for the indicated periods . For histological analyses , liver fragments were fixed in 4% formaldehyde for 24 h prior to paraffin embedding . Liver sections were 5 µm thick , and were stained according to the European Mouse Phenotyping Resource of Standardised Screens ( http://www . eumorphia . org ) . Liver fragments were also frozen for RNA and proteins extraction . Blood glucose levels were measured with One Touch Ultra Glucose meter ( LifeScan Inc , Milpitas , California ) . Triglycerides , free fatty acids and β-hydroxybutyrate measurement from serum were performed with an OLYMPUS AU-400 automated laboratory workstation . Insulin and glucagon were measured using Millipore Milliplex Kit and Glucagon quantikine Elisa kit ( R&D Systems Europ ) , respectively . Lactate and Pyruvate levels were measured using Abcam kits ab65331 and ab65342 , respectively . For the Pyruvate Tolerance Tests , overnight fasted ( 16 h ) mice were injected IP with pyruvate ( 2 g/kg ) , and the glucose level was measured before and after injection at the different time points . Immunostainings on liver sections ( 5 µm ) were performed using polyclonal anti-G6Pase ( ab83690 , Abcam ) and anti-PEPCK ( sc-74823 , Santa Cruz ) antibodies . As secondary antibody , Alexa Fluor Goat Anti-Rabbit IgG ( Invitrogen ) diluted 1∶200 was used . Counterstaining was performed with DAPI . Hepatocytes were isolated from liver tissues derived from E14 . 5-murine WT and TTD embryos as previously described [51] , [52] . After careful selection , WT and TTD hepatocytes were grown in Dulbecco's modified Eagle medium ( DMEM ) containing 10% of fetal calf serum ( FCS ) , 40 µg/ml gentamicin , 1x nonessential amino acid solution ( Fischer Scientific ) , 2 mM L-glutamine ( Invitrogen ) , 1x Insulin-Transferin-Selenium solution ( ITS , Invitrogen ) , 0 , 1 µg/ml dexamethasone ( Sigma Aldrich ) and antibiotic/antimycotic solution ( Invitrogen ) . Before stimulation of the gluconeogenesis pathway , cells were preincubated with 1 g/l glucose and red phenol-free medium containing 2 , 5% charcoal treated FCS and ITS for 16 hr . Hepatocytes were then treated in absence of glucose with 1 mM pyruvate , 10 µM Forskolin ( Sigma Aldrich ) and 150 nM glucagon ( Sigma Aldrich ) . 1×106 harvested cells were used to measure intracellular Glucose 6-Phosphate and NAD and NADH levels ( using Abcam assay kits ab83426 and ab65348 , respectively ) . Glucose output in the medium was measured with glucose assay kit ( Abcam ab65333 ) . Hepatocytes were transiently transfected with 1 µg of PGC-1α expression vector using the transfection reagent JetPEI ( Polyplus Transfection ) ; the cells were harvested at 72 h posttransfection . To potentiate the PGC-1α acetylation , WT hepatocytes were cotransfected with p300 and treated with nicotinamide ( 10 mM ) for 16 h before harvesting [16] . Total RNAs ( 2 µg ) were reverse-transcribed with Moloney murine leukemia virus RT ( Invitrogen ) using random hexanucleotides . Real-time quantitative PCR was done using the “FastStart DNA Master SYBR Green” kit and the Lightcycler apparatus ( Roche Diagnostic ) . The mRNA levels of interest were normalized to the 18S RNA amount , which was not affected by the different fasting periods ( in the liver ) and the pyruvate treatment ( in the hepatocytes ) . ChIP experiments were performed as previously described [21] . Primers were designed to amplify the proximal promoter region of PEPCK ( −300 to +13 ) , G6Pase ( −122 to +54 ) and PGC1α ( −134 to +19 ) . Monoclonal antibodies against the TFIIH subunits , RNA pol II , GST-Tag and Flag-Tag were produced at the IGBMC . Antibodies against PEPCK ( sc-74823 , Santa Cruz SC ) , G6Pase ( Ab83690 , AB Cam LTD ) , PGC-1α ( 4C1 . 3 , Calbiochem; AB3242 , Millipore ) , SIRT1 ( 2028 , Cell Signaling Technology CST and 3H10 . 2 Millipore , for murine and human SIRT1 , respectively ) , HNF-4α ( C-19 , SC ) , FOXO1 ( L27 , CST ) , acetylated FOXO1 ( D-19 , SC ) , CREB ( 48H2 , CST ) , β-tubulin ( KMX-1 , Millipore ) and acetylated lysine ( 9441 , SCT ) were purchased . Full-length human SIRT1 and mouse PGC-1α were obtained from Addgene ( plasmid #13735 and #1026 , respectively ) . Flag-PGC-1α was overexpressed and immunoprecipitated from cells . Truncated PGC-1α ( 36-797 amino acids , for solubility reasons ) [33] , was cloned into the bacterial expression vector pGEX-4T3 and purified via the GST-Tag to perform in vitro analyses . His-SIRT1 , GST-PGC-1α and GST-CTD proteins were produced in E . Coli strain BL21 and purified on either NI-NTA agarose ( Qiagen ) or glutathione column ( Thermo Scientific ) . Equal amounts ( 1 µg ) of recombinant proteins were incubated with highly purified TFIIH ( rIIH ) in the presence of [γ-32P] ATP ( 0 . 14 µM ) . Sf9 cells were infected with baculoviruses encoding either each Flag-subunit of TFIIH separately or all subunits together to produce an entire TFIIH . Whole extracts from infected cells were incubated with antibody against the Flag-Tag to immunoprecipitate either the entire TFIIH ( containing only the Flag-tagged XPB subunit ) or each Flag-tagged subunit individually . After incubation with His-SIRT1 and/or GST-PGC-1α during 2 h at 4°C and extensive washings ( 400 mM KCl ) , bound protein were resolved by SDS-PAGE and revealed by immunoblotting . | In eukaryotes , the expression of genes encoding proteins requires the action of hundreds of factors , together with the RNA polymerase II . While these factors are timely and selectively required for the expression of a given gene , little is known about their partnership upon gene expression . Our results reveal a cooperation between different types of transcription factors , namely the general transcription factor TFIIH , the cofactor PGC-1α and the deacetylase SIRT1 . Such partnership is however impaired when TFIIH is mutated , as observed in Trichothiodystrophy patients that develop premature ageing . These results thus shed light on the coordinated action of factors during transcription and allow us to better understand molecular deficiencies observed in many human diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetic",
"disorders",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"genetics",
"of",
"disease",
"dna",
"transcription"
] | 2014 | Dynamic Partnership between TFIIH, PGC-1α and SIRT1 Is Impaired in Trichothiodystrophy |
Cardiovascular diseases ( CVD ) and type 2 diabetes ( T2D ) are closely interrelated complex diseases likely sharing overlapping pathogenesis driven by aberrant activities in gene networks . However , the molecular circuitries underlying the pathogenic commonalities remain poorly understood . We sought to identify the shared gene networks and their key intervening drivers for both CVD and T2D by conducting a comprehensive integrative analysis driven by five multi-ethnic genome-wide association studies ( GWAS ) for CVD and T2D , expression quantitative trait loci ( eQTLs ) , ENCODE , and tissue-specific gene network models ( both co-expression and graphical models ) from CVD and T2D relevant tissues . We identified pathways regulating the metabolism of lipids , glucose , and branched-chain amino acids , along with those governing oxidation , extracellular matrix , immune response , and neuronal system as shared pathogenic processes for both diseases . Further , we uncovered 15 key drivers including HMGCR , CAV1 , IGF1 and PCOLCE , whose network neighbors collectively account for approximately 35% of known GWAS hits for CVD and 22% for T2D . Finally , we cross-validated the regulatory role of the top key drivers using in vitro siRNA knockdown , in vivo gene knockout , and two Hybrid Mouse Diversity Panels each comprised of >100 strains . Findings from this in-depth assessment of genetic and functional data from multiple human cohorts provide strong support that common sets of tissue-specific molecular networks drive the pathogenesis of both CVD and T2D across ethnicities and help prioritize new therapeutic avenues for both CVD and T2D .
Cardiovascular disease ( CVD ) and type 2 diabetes ( T2D ) are two leading causes of death in the United States [1] . Patients with T2D are at two to six times higher risk of developing CVD compared to those without T2D [2] , indicating the importance of targeting common pathogenic pathways to improve the prevention , diagnosis , and treatment for these two diseases . While decades of work has revealed dyslipidemia , dysglycemia , inflammation , and hemodynamic disturbances as common pathophysiological intermediates for both CVD and T2D [3–5] , very few studies have directly investigated the genomic architectures shared by the two diseases . While genetic factors are known to play a fundamental role in the pathogenesis of both CVD and T2D [6] , a direct comparison of the top risk variants between these diseases has revealed few overlapping loci in genome-wide association studies ( GWAS ) from multiple large consortia . Aside from the speculation that the strongest genetic risks may be disease-specific , the agnostic approach requiring the application of strict statistical adjustment for multiple comparisons also increases false negative rate because of the lack of “genome-wide significance” . To meet these challenges , we and others have previously shown that hidden disease mechanisms can be unraveled through the assessment of the combined activities of genetic loci with weak to moderate effects on disease susceptibility by integrating GWAS with functional genomics and regulatory gene networks [7–11] . Importantly , such high-level integration approaches are able to overcome substantial heterogeneity between independent datasets and extract robust biological signals across molecular layers , tissue types , and even species [8 , 12–14] . This advantage is mainly conferred by the aggregation of genetic signals from individual studies onto a comparable ground–molecular pathways and gene networks , before conducting meta-analysis across studies [14 , 15] . In other words , even if the genetic variants and linkage architecture can be different between studies , the biological processes implicated are more reproducible and comparable across studies [16] . In the current investigation , we employed a systematic data-driven approach that leveraged multi-dimensional omics datasets including GWAS , tissue-specific expression quantitative trait loci ( eQTLs ) , ENCODE , and tissue-specific gene networks ( Fig 1 ) . GWAS datasets were from three well-characterized and high-quality prospective cohorts of African Americans ( AA ) , European Americans ( EA ) , and Hispanic Americans ( HA ) —the national Women’s Health Initiative ( WHI ) [8] , the Framingham Heart Study ( FHS ) [17] , and the Jackson Heart Study ( JHS ) [18] . To maximize the reproducibility of our findings across different populations , we also incorporated meta-analyses of CVD and T2D genetics from CARDIoGRAMplusC4D [19] and DIAGRAM [20] . Further , we comprehensively curated functional genomics and gene networks derived from 25 tissue or cell types relevant to CVD and T2D . A streamlined integration of these rich data sources using our Mergeomics pipeline [14 , 15] enabled the identification of shared pathways , gene subnetworks , and key regulators for both CVD and T2D across cohorts and ethnicities . Finally , we validated the subnetworks using adipocyte and knockout mouse models , and confirmed their associations with cardiometabolic traits in the Hybrid Mouse Diversity Panel ( HMDP ) comprised of >100 mouse strains [21–23] .
We first investigated whether genetic risk variants of CVD and T2D from GWAS of each cohort/ethnicity were aggregated in a functionally coherent manner by integrating GWAS with tissue-specific eQTLs or ENCODE information and gene co-expression networks that define functional units of genes ( Fig 1A ) . Briefly , co-expression networks were constructed from an array of transcriptomic datasets of various tissues relevant to CVD and T2D ( details in Methods ) . These modules were mainly used to define sets of functionally related genes in a data-driven manner . Genes within the co-expression modules ( a module captures functionally related genes ) were mapped to single nucleotide polymorphisms ( SNPs ) that most likely regulate gene functions via tissue-specific eQTLs or ENCODE information . SNPs were filtered by linkage disequilibrium ( LD ) and then a chi-square like statistic was used to assess whether a co-expression module shows enrichment of potential functional disease SNPs compared to random chance using the marker set enrichment analysis ( MSEA ) implemented in our Mergeomics pipeline ( details in Methods ) [14] . Subsequently , meta-analyses across individual MSEA results at the co-expression module level were conducted using the Meta-MSEA function in Mergeomics to retrieve robust signals across studies . Among the 2 , 672 co-expression modules tested , 131 were found to be significant as defined by false discovery rate ( FDR ) < 5% in Meta-MSEA across studies ( Table 1 , S1 Table ) . Moreover , the majority of the disease relevant tissues or cell types included in our analysis yielded informative signal , supporting the systemic pathogenic perturbations known for CVD and T2D ( S1 Fig ) . Of the significant modules identified , 79 were associated with CVD and 54 with T2D . Two modules were associated with both diseases , with one enriched for “carbohydrate metabolism” genes and the other over-represented with “other glycan degradation; known T2D genes” ( Fig 2A , S1 Table ) . Examination of these two shared modules showed that the genetic signals driving the module significance were largely different between CVD and T2D , with 14 . 8% lead SNPs overlapping for the carbohydrate metabolism module and 5 . 8% lead SNPs overlapping for the glycan degradation module between diseases . These results indicate that the GWAS signals for the two diseases in each module do not necessarily overlap , but the CVD and T2D genes are likely functionally connected since they are co-expressed in the same modules and annotated with coherent functions . Additionally , the majority of the CVD modules and T2D modules were identified in more than one ethnic group based on MSEA analysis of individual studies , supporting consistency across ethnicities ( Fig 2B ) . Apart from the two directly overlapping modules , between the CVD- and T2D-associated modules there were many overlapping genes , indicating additional shared functions that contribute to both diseases ( S2 Fig ) . Upon annotating the disease-associated modules using functional categories curated in Kyoto Encyclopedia of Genes and Genomes ( KEGG ) and Reactome while correcting for the overlaps between pathways ( method details in S1 Text; S3 Fig; S2 Table ) , we found significant functional overlaps between the CVD and T2D modules ( overlap p = 3 . 1e-15 by Fisher’s exact test , Fig 2C ) . We further ranked all the enriched functional categories by the number of CVD/T2D modules that were annotated with each functional term ( Fig 3 ) , which showed a wide spectrum of biological processes shared by both CVD and T2D across ethnicities and cohorts . Of the top ranked processes for the significant co-expression modules identified , we observed well-established pathogenic processes such as lipid and fatty acid metabolism [24] , glucose metabolism [25] , oxidation [26] , and cytokine signaling [27] . Pathways previously implicated mainly for T2D such as beta-cell function were also found to be shared for both CVD and T2D . Interestingly , our completely data-driven approach also identified extracellular matrix ( ECM ) and branched chain amino acids ( BCAA ) metabolism as top functional categories whose roles in the development of cardiometabolic disorders have only been implicated in recent experimental work [28–30] . Furthermore , our analysis also revealed under-appreciated processes involving the neuronal system and transport of small molecules . The coexpression networks used above mainly served to capture coexpression patterns between genes and to define data-driven gene sets that contain functionally related genes , but they do not carry detailed topology information on gene-gene regulatory relationships . To dissect the gene-gene interactions within and between the 131 disease-associated modules , and to identify key perturbation points shared for both CVD and T2D modules , we used the GIANT networks [31] and Bayesian networks ( BNs ) from 25 CVD and T2D relevant tissue and cell types , which provide detailed topological information on gene-gene regulatory relationships necessary for the downstream network analysis . The BNs used in our study were generated using similar sets of mouse and human gene expression datasets as used for the co-expression networks , but additionally incorporated genetic data to model causal gene regulatory networks , whereas the GIANT networks were derived based on independent gene expression datasets and protein interaction information . We included both types of gene regulatory networks to increase the coverage of functional connections between genes and only considered KDs identified in both to enhance the robustness of KD prediction . Specifically , all genes in each of the co-expression modules genetically associated with CVD or T2D as identified in our study were mapped onto the GIANT and BN graphical networks to identify KDs using the weighted key driver analysis ( wKDA ) implemented in Mergeomics [14] , where KDs were defined as genes whose local network neighborhoods demonstrate significant enrichment of genes from disease-associated modules ( details in Methods; concept depicted in S4 Fig ) . Of note , wKDA gives higher weight to network edges that are consistent across network models constructed from independent studies , therefore alleviating potential bias caused by dataset heterogeneity . We identified 226 KDs that were consistently captured in Bayesian and GIANT network at Bonferroni-corrected p-value < 0 . 05 ( Fig 1B ) , among which 162 were KDs for both CVD and T2D associated modules . Bonferroni-correction was used here to focus on the strongest KDs for prioritization purposes . To further prioritize these 162 shared KDs , tissue-specific subnetworks of these KDs were evaluated using Meta-MSEA to rank the magnitude of their genetic association with CVD and T2D across cohorts , yielding 15 top-ranked KDs at FDR<10% in Meta-MSEA for CVD and T2D separately ( combined FDR<1% for both diseases simultaneously ) ( Fig 1B , Table 2 ) . The top KD subnetworks were related to similar pathogenic processes highlighted in the previous section , including cholesterol biosynthesis , respiratory electron transport , immune system and ECM . We further inferred the directionality of the effects of each specific KD on both diseases using GWAS signals mapped to each KD based on eQTLs or chromosomal distance ( details in Methods; results in S5 Fig ) . This analysis differentiated the KDs into those showing consistent direction of association for both CVD and T2D ( ACLY , CAV1 , SPARC , COL6A2 , IGF1 ) , inverse directions with CVD and T2D ( HMGCR , IDI1 ) , and uncertain directions ( Table 2 ) . Therefore , the shared KDs do not necessarily affect the risks for the two diseases in the same direction . The KDs and subnetworks were identified based on the full spectrum of genetic evidence ( from strong to moderate and subtle ) from the various GWAS datasets examined in the current study . To assess whether the top KD subnetworks were enriched for previously known disease genes that mostly represent the strong and replicated genes as a means of cross-validation , we manually curated previously reported genes associated with CVD , T2D , and intermediate metabolic traits related to CVD , T2D ( glucose , insulin , lipids , obesity ) from DisGeNET [32] and the NHGRI GWAS Catalog [6] ( Fig 1C , genes listed in S3 Table ) . The connection between the top 15 KDs and known genes for CVD , T2D and relevant cardiometabolic traits was confirmed by the significant over-representation of the known disease genes in KD subnetworks , with fold enrichment as large as 8 , confirming the strong biological importance of these KDs ( Fig 4A ) . Further , the top 15 KDs showed direct connections to 28 GWAS hits reaching genome-wide significance ( p < 5e-8 ) for CVD and 16 for T2D , which account for 35% ( fold = 3 . 35 , p = 7 . 18e-10 ) and 22% ( fold = 2 . 16 , p = 8 . 08e-4 ) of all reported significant GWAS signals for CVD and T2D in GWAS catalog , respectively . Two of the 15 top KDs , namely HMGCR and IGF1 , were previously identified as signals of genome-wide significance for obesity , lipids and T2D , all risk factors of CVD . Additionally , network visualization revealed tissue-specific KDs and interactions of CVD and T2D genes in many disease-relevant tissues including adipose , adrenal gland , artery , blood , digestive tract ( small intestine , colon ) , hypothalamus , islet , liver , lymphocyte , skeletal muscle , thyroid , and vascular endothelium ( Fig 4B ) . PCOLCE represents an intriguing hypothalamus KD that interacts with important energy homeostasis genes like leptin receptor LEPR , suggesting a role of neurohormonal control in CVD and T2D pathogenesis . In contrast , CAV1 appeared to interact extensively with other KDs in peripheral tissues , especially in the adipose tissue . CAV1 is a robust KD for CVD- and T2D-associated modules across multiple tissues , with the adipose tissue subnetwork of CAV1 containing the largest number of neighboring genes ( Fig 4B ) . In addition , adipose tissue is the only tissue where CAV1 is a KD in both the Bayesian networks and GIANT networks . These lines of evidence implicate the potential importance of CAV1 adipose subnetwork in the shared pathogenesis for both diseases . Indeed , Cav1-/- mice have been shown to alter the lipid profile , susceptibility to atherosclerosis , and insulin resistance [33 , 34] . To assess whether perturbation of this potential KD induces changes in the subnetwork genes as predicted by our network modeling , we performed validation by conducting siRNA-mediated knock down of Cav1 in differentiating mouse 3T3-L1 adipocytes and by evaluating the whole transcriptome alteration in mouse gonadal adipose tissue between wild type and Cav1-/- mice [33] ( Fig 1C; details in Methods ) . Of the 12 adipose network neighbors of Cav1 that were tested in vitro , 6 exhibited significant changes in expression level on day 2 after ~60% Cav1 knockdown using two siRNAs against Cav1 . In contrast , none of the 5 negative controls , which were randomly selected among adipocyte genes that are not connected to Cav1 or its first level neighbors in the adipose network , were affected after Cav1 perturbation ( Fig 5A ) . Cav1 knockdown also led to decreased expression of Pparg , a major adipocyte differentiation regulator ( S6 Fig ) , supporting a role of Cav1 in adipocyte differentiation as previously observed [35] . In 3-month-old Cav1-/- mice which showed perturbed lipid and insulin sensitivity profiles , we observed 1 , 474 differentially expressed genes ( DEGs ) at FDR<1% . We found that the first and second level neighbors of CAV1 in our predicted subnetwork showed significant enrichment for DEGs in adipose tissue induced by Cav1 knockout , with the degree of fold enrichment increasing as the statistical cutoff used to define DEGs became more stringent ( Fig 5B; subnetwork view with DEGs in S7 Fig ) . On the contrary , the third and fourth level neighbors of CAV1 in our predicted subnetwork did not exhibit such enrichment of DEGs ( Fig 5B ) . These experimental findings support that CAV1 is a key regulator of the subnetwork and the network structure predicted by our network modeling is reliable , although it is difficult to discern whether the network changes are related to alterations in adipocyte differentiation status . We also observed strong enrichment for the focal adhesion pathway in both the predicted Cav1 adipose subnetwork ( p = 9 . 6e-14 by Fisher’s exact test , fold enrichment = 6 . 0 ) and the differential adipose genes in Cav1-/- mice ( p = 6 . 9e-9 , fold enrichment = 3 . 5 ) . We further assessed the transcriptomic profiling in adipose ( relevant to T2D and CVD ) and aorta tissue ( main site of CVD ) in relation to 7 cardiometabolic phenotypes including adiposity , lipid levels ( triglyceride , LDL , HDL ) , fasting glucose , fasting insulin and HOMA-IR , across >100 mouse strains in two HMDP panels [21–23] . HMDP is a systems genetics resource that comprises more than 100 commercially available mouse strains differing in genetic composition , and has emerged as a power tool to study complex human diseases [22 , 36] . The biological relevance of HMDP to human pathophysiology has been reproducibly demonstrated [37–39] . Moreover , HMDP data was completely independent of the human-focused genetic datasets and the network datasets used in our primary integrative analysis ( Fig 1C ) . Here we selected two specific HMDP panels , high-fat ( HF ) and atherogenic ( ATH ) , in which mice were either fed with a high-fat high-sucrose diet or underwent transgenic expression of human APOE-Leiden and CETP gene as a pro-atherogenic background , respectively . These two panels were chosen for their representativeness of human T2D ( the HF panel ) and CVD ( the ATH panel ) pathology . First , we investigated the correlation between the expression of 14 top KDs ( no probe for KD MSMO1 in HMDP ) and cardiometabolic traits in the adipose and aorta tissues assessed in HMDP . All 14 KDs displayed significant trait association in HMDP , with the association for 11 KDs replicated in both the HF and ATH HMDP panels ( Fig 6A ) . Next , we retrieved the adipose and aorta gene-trait correlation statistics for the top KD subnetwork genes , and used MSEA to test whether genes in the KD subnetworks displayed an overall overrepresentation of strong trait association in HMDP . Again , the 14 KD subnetworks showed significant trait association after Bonferroni correction ( Fig 6B ) . These findings support that the close involvement of the KDs in cardiometabolic trait perturbation we predicted based on human datasets can be cross-validated in mouse models . Cav1 knockout in mice led to dysreuglation of the predicted subnetwork ( Fig 5B ) and significant alterations in cardiometabolic phenotypes [33 , 34] , supporting the causal role of CAV1 in both CVD and T2D . To further investigate the potential causal role of the top KDs and their subnetworks in CVD and T2D , we conducted integrative analysis of the KD subnetworks to assess their disease association using GWAS results for the 7 cardiometabolic traits from HMDP and tissue-specific cis-eQTLs ( Fig 1C ) . By mapping GWAS signals to genes using adipose or aorta eQTLs and testing for enrichment of genetic association with cardiometabolic traits within the KD subnetwork genes using MSEA , we found consistent and significant association between cardiometabolic traits and the subnetworks of KDs ACAT2 , CAV1 , COL6A2 , IGF1 , PCOLCE , and SPARC across adipose and aorta ( Fig 6C ) . These results informed by mouse GWAS support a potential causal role of these top KDs in perturbing gene networks in multiple tissues to trigger CVD and T2D .
CVD and T2D are highly correlated complex diseases and share many common risk factors . Multiple genetic variants may individually exert subtle to strong effects on disease pathogenesis , and in aggregate perturb diverse pathogenic pathways [8 , 9 , 13 , 19 , 20 , 40] . In this systems-level , data-driven analysis of GWAS from several large and high-quality cohorts of diverse ethnicities , integrated with functional data ( from ENCODE , eQTLs , tissue-specific co-expression and regulatory networks constructed from human and mouse experiments ) , we identified both known and novel pathways and gene subnetworks that were genetically linked to both CVD and T2D across cohorts and ethnicities . Further , KDs in tissue-specific subnetworks appear to regulate many known disease genes for increased risk of CVD and T2D . Lastly , we experimentally validated the network topology using in vitro adipocyte and data from in vivo gene knockout models , and confirmed the role of the top KDs and subnetworks in both CVD and T2D traits in independent sets of mouse studies . The data-driven nature of the current study offers several strengths . First , we incorporated the full-scale of genetic variant-disease association from multiple cohorts , ethnicities and disease endpoints , allowing for the detection of subtle to moderate signals , as well as comparison and replication of results across diseases and populations . More importantly , by focusing on results that demonstrate consistent significance at pathway and network level , we overcome the difficulties in harmonizing independent datasets that are complicated by substantial heterogeneity due to platform differences and population substructure . This is because disease signals across populations are more conserved at pathway level than at individual variant and gene levels [12 , 14 , 16] . Second , the comprehensive incorporation of tissue-specific eQTLs , coupled with the use of tissue-specific networks , enhances our ability to achieve better functional mapping between genetic variants and genes , and uncover systems-level regulatory circuits for CVD and T2D in a tissue-specific fashion . Third , data-driven modules and networks used in this study increase the potential for novel discovery as gene-gene interactions are defined by data rather than prior knowledge . As the network models were from many independent studies reflecting diverse physiological conditions , leveraging these datasets and network models offers more comprehensive coverage of biological interactions than any given dataset can provide and has proven a valuable approach to unveil novel biological insights [9 , 13 , 41] . While some of our findings confirmed those from previous canonical pathway-based analysis on disease processes including ECM-receptor interaction and cell-adhesion , and KDs such as SPARC [8] , our data-driven approach in the current study uncovered numerous novel genes , pathways , and gene subnetworks . A likely reason for the enhanced discovery potential of co-expression modules is that several interacting pathways could be co-regulated in a single module , or a pathway could interact with other poorly annotated processes in a module to together confer disease risk . The use of modules capturing such interactions improves the statistical power , in contrast to testing the pathways individually . Lastly , we conducted cross-validation studies in support of the functional roles of specific KDs and subnetworks in CVD and T2D using independent experimental models . We acknowledge the following limitations in our study . First , our analyses were constrained by the coverage of functional datasets that are currently available , which causes uneven tissue coverage between data types and statistical bias towards more commonly profiled tissues such as adipose and liver , making it difficult to achieve precise inference for all relevant tissues . Although we believe this does not necessarily undermine the validity of the main findings from our study , we acknowledge that we likely have missed relevant biology from tissues with fewer studies and smaller sample sizes . Further investigation is needed when additional relevant datasets become available . Secondly , our FDR estimates in MSEA do not take into consideration the gene overlap structure among co-expression modules , due to the challenge in accurately adjusting for the various degrees of overlaps between module pairs . To alleviate this limitation , we focus on modules and pathways demonstrating consistency across datasets and merge overlapping modules subsequently . Thirdly , although we conducted validation experiments on the CAV1 subnetwork in both in vitro and in vivo models and cross-validated the importance of the predicted top key drivers and subnetworks in two independent large-scale mouse population studies , further experiments are warranted to thoroughly test the causality of the predicted KDs and elucidate the detailed tissue-specific mechanisms of the KDs on CVD and T2D . This is particularly important considering the limited overlaps in the modules and KDs identified from our study and the ones identified in two recent multi-tissue network analysis of cardiometabolic diseases [10 , 11] . Only 7 KDs overlapped including APOA1 , CD2 , CEBPD , CENPF , CSF1R , CTSS , UBE2S . Methodological differences in network inference and key driver analysis and differences in the pathophysiological conditions of the study populations could explain the discrepancies . Lastly , ethnic-specific and sex-specific mechanisms await future exploration . There are several direct implications that can be drawn from the results of our integrative analyses of both observational and experimental data . First , it appears that pathogenic pathways for CVD and T2D are indeed common in ethnically diverse populations . These shared pathways capture most of the critical processes that have been previously implicated in the development of either T2D or CVD , including metabolism of lipids and lipoproteins , glucose , fatty acids , bile acids metabolism , biological oxidation , coagulation , immune response , cytokine signaling , and PDGF signaling . Second , BCAA metabolism and ECM are among the top and common pathways identified . Our finding on BCAA is consistent with recent work relating serum levels of BCAA to risk of CVD and T2D in large prospective cohorts [42 , 43] , although whether BCAA is a “pathophenotype” or strong pathogenic factor has been debated [28 , 44] . Our findings support a causal role of BCAA because 1 ) both CVD and T2D risk variants were enriched in the co-expression modules related to BCAA degradation , and 2 ) 15 genes in the BCAA pathway were part of the top KD subnetworks , representing a significant enrichment of BCAA genes ( fold enrichment = 3 . 02 , Fisher’s exact test p = 1 . 4e-5 ) . Of note , BCAA genes themselves carry few genetic risk variants for CVD and T2D , albeit their network neighboring genes are highly enriched for disease variants , which may result from negative evolutionary pressure due to the critical role of BCAA . More recently , Jang and colleagues have shown BCAA catabolism can cause insulin resistance , providing further support for the causal role of BCAA for both CVD and T2D [45] . Our finding on the role of ECM in both CVD and T2D is also in line with recent reports [8 , 13 , 29 , 30 , 46] . In the top enriched subnetworks , ECM genes appear to exert strong effect ( Fig 4B ) coordinating other processes such as cholesterol metabolism , energy homeostasis , and immune response across a wide range of peripheral tissues and endocrine axis . This substantiates the importance of ECM modeling as a mechanistic driver for CVD and T2D . Secondly , our comprehensive network modeling identified critical disease modulators and key targets whose functional roles were subsequently supported by multiple cross-validation efforts . This supports the use of network modeling to unravel and prioritize promising top targets that may have high pathogenic potential for both CVD and T2D . The KDs we identified can be considered as “highly confident” for the following reasons: 1 ) they are KDs for both CVD and T2D associated modules , 2 ) the tissue-specific subnetworks of these KDs show significant and replicable association with both diseases , 3 ) their subnetworks are highly enriched with known CVD and T2D genes , 4 ) in vitro siRNA knockdown and in vivo knockout mouse experiments confirm the role of a central KD CAV1 in regulating the downstream genes as predicted in our network model , and 5 ) both the expression levels of KDs and the genetic variants mapped to the KD subnetworks are significantly associated with CVD and T2D relevant traits in independent mouse populations with naturally occurring genetic variations . Thirdly , most KDs are not GWAS signals reaching genome-wide significance , nor are they rare-variant carrying genes , indicating that standard genetic studies miss important genes that orchestrate known CVD and T2D genes . The phenomenon may reflect a negative evolutionary pressure experienced by the KDs due to their crucial functions . In support of this hypothesis , we found a significant enrichment of human essential genes lacking functional variations among the 162 KDs identified in our study [47] ( Fold = 1 . 41 , p = 9 . 02e-3 ) . This is consistent with previous findings [8 , 9 , 13] reaffirming the power and reliability of our approach in uncovering hidden biological insights particularly in a systematic integrative manner . The connections between KDs and other disease genes revealed by our study warrant future investigation into the potential gene-gene interactions . Indeed , a closer examination of the biological functions from the top shared KDs further corroborates their disease relevance . For instance , our network modeling identified HMGCR as a top KD , consistent with its primary role as the target for cholesterol-lowering HMG-CoA inhibitors , namely statins . Our directionality inference analysis indicates that HMGCR is associated with CVD and T2D in opposite directions . This is consistent with the recent findings that genetic variations in HMGCR that decrease CVD risk cause slightly increased T2D risk , and statin drugs targeting HMGCR reduces CVD risk but increases T2D risk [48–50] . CAV1 and IGF1 represent two tightly connected multi-functional KDs . CAV1 null mice were found to have abnormal lipid levels , hyperglycemia , insulin resistance and atherosclerosis [33 , 34] . Consistent with these observations , we found strong association of CAV1 expression levels as well as CAV1 network with diverse cardiometabolic traits in both human studies and mouse HMDP panels . Our data-driven approach also revealed the central role of CAV1 in adipose tissue by elucidating its connection to a large number of CVD and T2D GWAS genes and to genes involved in focal adhesion and inflammation ( Fig 4 ) , which could drive adipocyte insulin resistance [51 , 52] . The regulatory effect of CAV1 on neighboring genes was subsequently validated using in vitro adipocyte and in vivo mouse models . Moreover , our network modeling also captured the central role of CAV1 in muscle and artery tissues , suggesting multi-tissue functions of CAV1 in the pathogenic crossroads for CVD and T2D . The other multi-functional KD , IGF1 , is itself a GWAS hit for fasting insulin and HOMA-IR . Despite being primarily secreted in liver , in our study IGF1 demonstrated an adrenal gland and muscle specific regulatory circuit with CVD and T2D genes , suggesting that it may confer risk to these diseases through the adrenal endocrine function and muscle insulin sensitivity . The three ECM KDs we identified , SPARC , PCOLCE and COL6A2 , were especially interesting due to their consistent and strong impact on diverse cardiometabolic traits shown in our cross-validation analyses in HMDP ( Fig 4 , Fig 6 ) . SPARC encodes osteonectin , which is primarily circulated by adipocytes . It inhibits adipogenesis and promotes adipose tissue fibrosis 50 . SPARC is also associated with insulin resistance and coronary artery lesions 51 , 52 . PCOLCE ( procollagen C-endopeptidase enhancer ) represents a novel regulator for hypothalamus ECM that could potentially disrupt the neuroendocrine system . COL6A2 , on the other hand , provides new insights into how collagen may affect cardiometabolic disorders: in adrenal tissue COL6A2 is connected to IGF1R , the direct downstream effector for KD IGF1 . Importantly , our directionality analysis suggests that while some KDs such as CAV1 may have similar directional effects on CVD and T2D , cases like HMGCR that show opposite effects on these diseases are also present . Therefore , it is important to test the directional predictions to prioritize targets that have the potential to ameliorate both diseases and deprioritize targets with opposite effects on the two diseases . In summary , through integration and modeling of a multitude of genetics and genomics datasets , we identified key molecular drivers , pathways , and gene subnetworks that are shared in the pathogenesis of CVD and T2D . Our findings offer a systems-level understanding of these highly clustered diseases and provide guidance on further basic mechanistic work and intervention studies . The shared key drivers and networks identified may serve as more effective therapeutic targets to help achieve systems-wide alleviation of pathogenic stress for cardiometabolic diseases , due to their central and systemic role in regulating scores of disease genes . Such network-based approach represents a new avenue for therapeutic intervention targeting common complex diseases .
Detailed GWAS information including sample size , ethnicity and genotyping platform was described in S4 Table and S1 Text . Briefly , p-values of qualified single nucleotide polymorphisms ( SNPs ) at minor allele frequency > 0 . 05 and imputation quality > 0 . 3 for CVD and T2D were collected for all available GWAS datasets ( WHI-SHARe , WHI-GARNET , JHS , FHS , CARDIoGRAMplusC4D [19] , and DIAGRAM [20] ) . SNPs meeting the following criteria were further filtered out: 1 ) ranked in the bottom 50% ( weaker association ) based on disease association p-values and 2 ) in strong linkage disequilibrium ( LD ) ( r2 > 0 . 5 ) according to ethnicity-specific LD data from Hapmap V3 [53] and 1000 Genomes[54] . For each GWAS dataset , LD filtering was conducted by first ranking SNPs based on the association p values and then checking if the next highest ranked SNP was in LD with the top SNP . If the r2 was above 0 . 5 , the SNP with lower rank was removed . The step was repeated by always checking if the next SNP was in LD with any of the already accepted ones . Using the Weighted Gene Co-expression Network Analysis ( WGCNA ) [55] , we constructed gene co-expression modules capturing significant co-regulation patterns and functional relatedness among groups of genes in multiple CVD- or T2D-related tissues ( including aortic endothelial cells , adipose , blood , liver , heart , islet , kidney , muscle and brain ) obtained from nine human and mouse studies ( S5 Table ) . Modules with size smaller than 10 genes were excluded to avoid statistical artifacts , yielding 2 , 672 co-expression modules . These coexpression modules were used as a collection of data-driven sets of functionally connected genes for downstream analysis . The potential biological functions of each module were annotated using pathway databases Reactome and KEGG , and statistical significance was determined by Fisher’s exact test with Bonferroni-corrected p< 0 . 05 . eQTLs establish biologically meaningful connections between genetic variants and gene expression , and could serve as functional evidence in support of the potential causal role of candidate genes in pathogenic processes[56 , 57] . We therefore conducted comprehensive curation for significant eQTLs in a total of 19 tissues that have been identified by various consortia ( including the Genotype-Tissue Expression ( GTEx ) [58] , Muther [59] and Cardiogenics [60] , and additional independent studies; S6 Table ) . Additional functional genomics resources from ENCODE were also curated to complement the eQTLs for SNP-gene mapping ( S1 Text ) . MSEA was used to identify co-expression modules with over-representation of CVD- or T2D-associated genetic signals for each disease GWAS in each cohort/ethnicity in a study specific manner . MSEA takes into three input: 1 ) Summary-level results of individual GWAS ( WHI , FHS , JHS , CARDIoGRAM+C4D , DIAGRAM ) for the LD-filtered SNPs; 2 ) SNP-gene mapping information , which could be determined by tissue-specific cis-eQTLs , ENCODE based functional annotation and chromosome distance based annotation . Cis-eQTLs is defined as eQTLs within 1MB of the transcription starting sites of genes . For ENCODE , we accessed the Regulome database and used the reported functional interactions to map SNPs to genes by chromosomal distance . Only SNPs within 50kb of the gene region and have functional evidence in Regulome database were kept; 3 ) Data-driven co-expression modules from multiple human and mouse studies as described above . Tissue-specificity was determined by the tissue origins of eQTLs and ethnic specificity was determined by the ethnicity of each GWAS cohort , since the human disease genetic signals and human eQTL mapping were the main driving factors to determine the significance of the modules . MSEA employs a chi-square like statistic with multiple quantile thresholds to assess whether a co-expression module shows enrichment of functional disease SNPs compared to random chance [14] . The varying quantile thresholds allows the statistic to be adoptable to studies of varying sample size and statistical power . For the list of SNPs mapped to each gene-set , MSEA tested whether the SNP list exhibited significant enrichment of SNPs with stronger association with disease using a chi-square like statistic: χ=∑i=1nOi−EiEi+κ , where n denotes the number of quantile points ( we used ten quantile points ranging from the top 50% to the top 99 . 9% based on the rank of GWAS p values ) , O and E denote the observed and expected counts of positive findings ( i . e . signals above the quantile point ) , and κ = 1 is a stability parameter to reduce artefacts from low expected counts for small SNP sets . The null background was estimated by permuting gene labels to generate random gene sets matching the gene number of each co-expression module , while preserving the assignment of SNPs to genes , accounting for confounding factors such as gene size , LD block size and SNPs per loci . For each co-expression module , 10000 permuted gene sets were generated and enrichment P-values were determined from a Gaussian distribution approximated using the enrichment statistics from the 10000 permutations and the statistics of the co-expression module . Finally , Benjami-Hochberg FDR was estimated across all modules tested for each GWAS . To evaluate a module across multiple GWAS studies , we employed the Meta-MSEA analysis in Mergeomics , which conducts module-level meta-analysis to retrieve robust signals across studies . Meta-MSEA takes advantage of the parametric estimation of p-values in MSEA by applying Stouffer’s Z score method to determine the meta-Z score , then converts it back to a meta P-value . The meta-FDR was calculated using Benjamini-Hochberg method . Co-expression modules with meta-FDR < 5% were considered significant and included in subsequent analyses . We used graphical gene-gene interaction networks including the GIANT networks [31] and Bayesian networks ( BN ) from 25 CVD and T2D relevant tissue and cell types ( S7 Table , S1 Text ) to identify KDs . If more than one dataset was available for a given tissue , a network was constructed for each dataset and all networks for the same tissue were combined as a union to represent the network of that tissue , with the consistency of each network edge across datasets coded as edge weight . The co-expression modules genetically associated with CVD or T2D identified by Meta-MSEA were mapped onto these graphical networks to identify KDs using the weighted key driver analysis ( wKDA ) implemented in Mergeomics [14] . wKDA uniquely consider the edge weight information , either in the form of edge consistency score in the case of BNs or edge confidence score in the case of GIANT networks . Specifically , a network was first screened for suitable hub genes whose degree ( number of genes connected to the hub ) is in the top 25% of all network nodes . Once the hubs have been defined , their local one-edge neighborhoods , or “subnetworks” were extracted . All genes in each of the CVD and T2D-associated gene sets that were discovered by meta-MSEA were overlaid onto the hub subnetworks to see if a particular subnetwork was enriched for the genes in CVD/T2D associated gene sets . The edges that connect a hub to its neighbors are simplified into node strengths ( strength = sum of adjacent edge weights ) within the neighborhood , except for the hub itself . The test statistic for the wKDA is analogous to the one used for MSEA: χ=O−EE−κ , except that the values O and E represent the observed and expected ratios of genes from CVD/T2D gene sets in a hub subnetwork . In particular , E=NkNpN is estimated based on the hub degree Nk , disease gene set size Np and the order of the full network N , with the implicit assumption that the weight distribution is isotropic across the network . Statistical significance of the disease-enriched hubs , henceforth KDs , is estimated by permuting the gene labels in the network for 10000 times and estimating the P-value based on the null distribution . To control for multiple testing , stringent Bonferroni adjustment was used to focus on the top robust KDs . KDs shared by CVD and T2D modules are prioritized based on the following criteria: i ) Bonferroni-corrected p< 0 . 05 in wKDA , ii ) replicated by both GIANT networks and Bayesian networks , and iii ) the genetic association strength between the KD subnetworks ( immediate network neighbors of the KDs ) and CVD/T2D in Meta-MSEA . Finally , Cytoscape 3 . 3 . 0 [61] was used for disease subnetwork visualization . We used the genetic effect direction of KDs as a proxy for probable effect direction of the KD subnetworks . For each KD , we retrieved their tissue-specific eQTLs as well as variants within 50kb of the gene region , whose genetic association information was available in both CARDIoGRAMplusC4D and DIAGRAM , the two large meta-consortia of GWAS for CVD and T2D . CVD/T2D association beta-values of mapped variants of KDs were then extracted , and the signs of beta-values were examined to ensure they were based on the same reference alleles between GWAS . Lastly , for all mapped variants on each KD , the signs of the beta-value for CVD and T2D were compared and statistical significance of the proportion of variants with similar or opposite effect direction between diseases was determined by z-test . We chose to validate the predicted adipose subnetwork of a top ranked KD of both CVD and T2D , Cav1 , in 3T3-L1 adipocytes . Cells were cultured to confluence and adipocyte differentiation was induced using MDI differentiation medium ( S1 Text ) . Two days after the initiation of differentiation , cells were transfected with 50 nM Cav1 siRNAs ( 3 distinct siRNAs were tested and two of the strongest ones were chosen ) or a scrambled control siRNA . For each siRNA , two separate sets of transfection experiments were conducted , with three biological replicates in each experiment . Two days after transfection , cells were collected for total RNA extraction , reverse transcription and quantitative PCR measurement of 12 predicted Cav1 subnetwork genes and 5 random genes not within the subnetwork as negative controls ( S1 Text ) . β-actin was used to normalize the expression level of target genes . We accessed the gonadal white tissue gene expression data of 3-month-old wild type and Cav1-/- male mice ( N = 3/group ) from Gene Expression Omnibus ( GEO accession: GSE35431 ) . Detailed description of the data collection procedures has been described previously [33] . Gene expression was profiled using Illumina MouseWG-6 v2 . 0 expression beadchip and normalized using robust spline . Differentially expressed genes ( DEGs ) between genotype groups were identified using linear model implemented in the R package Limma and false discovery rate was estimated using the Benjamini-Hochberg procedure [62] . DEGs at different statistical cutoffs were compared to CAV1 subnetwork genes at different levels ( i . e . , 1 , 2 , 3 , or 4 edges away from CAV1 ) to assess overlap and significance of overlap was evaluated using Fisher’s exact test . To further validate the role of KD subnetworks in CVD and T2D , we incorporated genetic , genomic and transcriptomic data from HMDP ( comprised of >100 mouse strains differing by genetic composition ) [21–23] . HMDP data was from two panels , one with mice fed with a high-fat diet ( HF-HMDP ) [22] , and the other with hyperlipidemic mice made by transgenic expression of human APOE-Leiden and CETP gene ( ATH-HMDP ) [23] . For HF-HMDP , we retrieved gene-trait correlation data for adipose tissue ( due to its relevance to both CVD and T2D ) and 7 core cardiometabolic traits including adiposity , fasting glucose level , fasting insulin level , LDL , HDL , triglycerides and homeostatic model assessment-insulin resistance ( HOMA-IR ) . For ATH-HMDP , we retrieved aorta gene-trait correlation ( aorta tissue is the main site for CVD in mice ) for all 7 traits . In addition to assessing the trait association strengths of individual KDs , we also used MSEA to evaluate the aggregate association strength of the top CVD/T2D KD subnetworks with the traits at both transcription and genetic levels through transcriptome-wide association ( TWAS ) and GWAS in HF-HMDP and ATH-HMDP ( S1 Text ) . | Cardiovascular disease ( CVD ) and type 2 diabetes ( T2D ) are two tightly interrelated diseases that are leading epidemics and causes of deaths around the world . Elucidating the mechanistic connections between the two diseases will offer critical insights for the development of novel therapeutic avenues to target both simultaneously . Because of the challenging complexity of CVD and T2D , involving numerous risk factors , multiple tissues , and multidimensional molecular alterations , few have attempted such an investigation . We herein report a comprehensive and in-depth data-driven assessment of the shared mechanisms between CVD and T2D by integrating genomics data from diverse human populations including African Americans , Caucasian Americans , and Hispanic Americans with tissue-specific functional genomics information . We identified shared pathways and gene networks informed by CVD and T2D genetic risks across populations , confirming the importance of well-established processes , as well as unraveling previously under-appreciated processes such as extracellular matrix , branched-chain amino acid metabolism , and neuronal system for both diseases . Further incorporation of tissue-specific regulatory networks pinpointed potential key regulators that orchestrate the biological processes shared between the two diseases , which were cross-validated using cell culture and mouse models . This study suggests potential new therapeutic targets that warrant further investigation for both CVD and T2D . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"cardiovascular",
"medicine",
"animal",
"models",
"diabetes",
"mellitus",
"model",
"organisms",
"endocrine",
"disorders",
"network",
"analysis",
"genome",
"analysis",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"type",
"2",
"diabetes",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"gene",
"mapping",
"endocrinology",
"mouse",
"models",
"molecular",
"biology",
"metabolic",
"disorders",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cardiovascular",
"diseases",
"genomics",
"genetics",
"of",
"disease",
"computational",
"biology",
"human",
"genetics"
] | 2017 | Shared genetic regulatory networks for cardiovascular disease and type 2 diabetes in multiple populations of diverse ethnicities in the United States |
Most DNA viruses replicate in the cell nucleus , although the specific sites of virion assembly are as yet poorly defined . Electron microscopy on freeze-substituted , plastic-embedded sections of murine polyomavirus ( PyV ) -infected 3T3 mouse fibroblasts or mouse embryonic fibroblasts ( MEFs ) revealed tubular structures in the nucleus adjacent to clusters of assembled virions , with virions apparently “shed” or “budding” from their ends . Promyelocytic leukemia nuclear bodies ( PML-NBs ) have been suggested as possible sites for viral replication of polyomaviruses ( BKV and SV40 ) , herpes simplex virus ( HSV ) , and adenovirus ( Ad ) . Immunohistochemistry and FISH demonstrated co-localization of the viral T-antigen ( Tag ) , PyV DNA , and the host DNA repair protein MRE11 , adjacent to the PML-NBs . In PML−/− MEFs the co-localization of MRE11 , Tag , and PyV DNA remained unchanged , suggesting that the PML protein itself was not responsible for their association . Furthermore , PyV-infected PML−/− MEFs and PML−/− mice replicated wild-type levels of infectious virus . Therefore , although the PML protein may identify sites of PyV replication , neither the observed “virus factories” nor virus assembly were dependent on PML . The ultrastructure of the tubes suggests a new model for the encapsidation of small DNA viruses .
Increasing evidence suggests that the assembly of many viruses occurs at specific intracellular sites , which have been termed “virus factories” [1] , [2] . These subcellular domains are gathering points for coordinating genome replication and capsid protein assembly into virions . The ultrastructure of the factories has been determined for a number of RNA viruses that assemble in the cytoplasm . For example , specific membrane compartments such as the ER appear to be co-opted and re-configured such that the surface of the membrane is used as a scaffold , where viral replication is spatially juxtaposed with capsid proteins specifically delivered to these locations [3]–[5] . Such scaffolds likely have a significant kinetic impact on virion assembly . Structural information concerning assembly sites for DNA viruses that replicate and assemble in the nucleus is less clear , perhaps due to the intrinsic complexity and dynamic nature of the nuclear architecture . Polyomaviruses are small non-enveloped dsDNA viruses that replicate and assemble in the cell nucleus . The viral capsid is comprised of 72 pentamers of the major capsid protein VP1 , each associated with a single copy of a minor capsid protein , either VP2 or VP3 [6] . A complex of VP1 pentamer , with VP2/3 , is nuclear imported as a capsid “subunit” late in infection [7] . Replication of the viral genome is accomplished through the interaction of the viral large T-antigen ( Tag ) and its recruitment of cellular DNA replication proteins [8] , [9] . How the capsid protein subunits specifically identify the viral genome is unknown , but sequences near the viral origin of replication along with the J-domain of large Tag both appear to have important functions [10]–[13] . In contrast to the larger DNA viruses , such as herpes [14] and Ad [15] , assembly of polyomavirus has been hypothesized to involve “polymerization” of the capsid protein subunits onto the viral minichromosomes rather than injection of the viral genome into a preformed capsid structure [16] . However , intermediates in the assembly process are not well defined . Candidate sites for nuclear virus assembly factories are promyelocytic leukemia nuclear bodies or PML-NBs ( previously termed ND10s , PML oncogenic domains , or PODs ) . Maul et al first associated PML-NBs with DNA virus replication , specifically studying infection by herpes simplex virus type-1 ( HSV-1 ) , adenovirus type 5 ( Ad5 ) , and SV40 [17]–[20] . PML-NBs are functionally heterogeneous intra-nuclear structures , operationally identified by immuno-staining with anti-PML protein antibodies ( i . e . , a subset of nuclear “dots” ) . They have been associated with such diverse functions as interferon antiviral responses , DNA damage repair , and the p53 response [21] , [22] . Although a large number of proteins have been found associated with PML-NBs , constitutive proteins include PML , which appears essential for PML-NB component protein co-localization , along with Sp100 and Daxx . PML-NB formation requires PML modification by sumoylation [23] . Using 4Pi fluorescence laser-scanning microscopy , PML-NBs have been modeled in three dimensions as spheres of varying diameters with a 50–100 nm thick shell comprised of PML and Sp100 proteins stabilized by sumoylation [24] . PML-NBs in uninfected cells may serve as scaffolds for assembling a variety of DNA repair and checkpoint signaling proteins in response to DNA damage , such as that caused by UV-irradiation [25] , [26] . Many DNA viruses appear to usurp the cellular DNA repair proteins for their replication [27] , and if replication and virion assembly are spatially coupled , then perhaps these viruses also utilize PML-NBs for capsid assembly . Polyomavirus ( SV40 , BKV , JCV ) DNA replication has been localized adjacent to PML-NBs [17] , [28]–[30] , and associated with recruitment of cellular DNA damage-related proteins , such as ATM kinase and the MRN complex [31] , [32] . In contrast to the large DNA viruses , infection with polyomaviruses does not grossly disrupt the morphology of PML-NBs , although their number and size may increase and alterations in their function may be detected [30] . Tag and VP1 have been co-localized with PML-NBs during BKV , SV40 and JCV infection supporting a spatial coupling of replication with assembly [28] , [29] . PML-NBs also are associated with papillomavirus ( PV ) replication [33]–[38] . Day et al have suggested a model in which the L2 protein first localizes to PML-NBs , associates with the viral E2 protein , which specifically labels the PV viral genome , and subsequently recruits the major capsid protein L1 to the genome for virion assembly [33] , [38] . However , neither E2-dependent transcription nor PV DNA replication have been found to be dependent on PML-NBs , and the PV L2 capsid protein does not require the PML protein itself for localization into specific foci or to form virus-like particles [37] , [39] . The convergence of Tag , DNA repair proteins , replicating viral DNA , and capsid proteins at PML-NBs during polyomavirus infection suggests that these sites may function as virus factories where all components in the assembly process interact . We have used electron microscopy and tomography of cryo-preserved samples to identify ultrastructural elements that might represent these factories in mouse polyomavirus ( PyV ) -infected mouse cells . Adjacent to growing virion clusters we identified tubular structures that appear to give rise to new virions . These structures were independent of the presence of the PML protein , and PML−/− cells and knockout mice were unaffected in virus replication . Moreover , the ultrastructure of the tubules suggests a new model for small DNA virus encapsidation .
We used electron microscopy and tomographic three-dimensional ( 3-D ) reconstructions of freeze-substituted , plastic-embedded specimens to identify possible sites and structural features of PyV virion assembly . PyV-infected mouse 3T3 fibroblasts were preserved by high-pressure freezing followed by low temperature freeze-substitution and resin embedding . PyV infection proceeds asynchronously and rapidly , consequently progressive stages of infection stages are observed in a single sample . Thus , the time course of infection was operationally-defined based upon the presence , size and number of virus clusters observed in the nucleus of an individual cell . Early in infection full virions were somewhat dispersed throughout the volume of the nucleus and not yet arranged in clusters , although they were closely associated with tubular structures ( Figure 1A and inset ) . The tubular structures were often located near the periphery of cellular condensed chromatin ( Figure 1A ) , consistent with previous observations for JCV [29] , [40] . As infection proceeded , the number of virions increased , forming virus clusters or arrays , remaining associated with the tubes ( Figure 1B , data not shown ) . The tubular structures were absent , or could not be detected , late in infection when arrays of progeny virions filled the nucleus ( Figure 1C ) . The temporal appearance of the tubes is consistent with their function as an assembly intermediate . When a montage of tomograms of Epon-embedded , semi-thick ( 300 nm ) serial sections from a PyV-infected 3T3 cell ( Figure 1B ) was modeled , the seemingly isolated virus clusters were actually interconnected by the tubular structures ( Figure 1D; see Supporting information Video S1 ) . Additionally , the sites of virion accumulation were no longer restricted to the periphery of the nucleus , but rather virus clusters and tubes were found throughout the interchromatin space ( Figure 1B ) . The tubular structures were not observed in uninfected cells indicating that they are virus-induced , nor were they present in cells still early during infection ( <30 hpi ) before VP1 expression ( data not shown ) . The diameter of the virions in these preparations was approximately 40 nm , which is consistent with shrinkage due to sample preparation . When the tubular structures were imaged in thin-sections ( 70 nm ) of plastic-embedded PyV-infected 3T3 cells , their average diameter was 35–40 nm , and two types could be distinguished based upon the presence or absence of an electron dense core ( Figure 2A ) . This distinction is similar to the differences in electron densities seen for empty versus full virus capsids ( Figure 2A ) , where full capsids display a dark , electron dense core due to the presence of DNA ( Figure 2A ) . In order to obtain a 3-D representation of the virus factories , a semi-thick ( 300 nm ) section was imaged as a dual-axis tilt-series ( ±60° in 1° increments , 200 kV ) in the electron microscope . The tomographic reconstructions of a single virus factory revealed that the virus cluster contained predominantly full virions , and the tubular structures exhibited a “full” and “empty” morphology similar to what was observed in thin sections and for virions ( Figure 2B ) . 3-D modeling of the virus factory revealed that the majority of the tubes were “full” and that their orientation varied with respect to the virus cluster ( Figure 2C; see Supporting information Video S2 ) . At high resolution , the virions appeared to have a dense core , representing the viral genome , surrounded by a lower density region . This outer density was consistent with the thickness of the capsid shell . The tubular structures had a remarkable substructure . In cryo-substituted , Epon-embedded preparations the tubes appeared to have a dense core surrounded by a lighter , uniform density similar to that seen for virions ( Figure 3A ) . Along the tubes “bubble-like” structures were observed ( Figure 3A inset ) similar to that observed in the tomogram of the single virus factory ( see Supporting information Video S2 ) . The ends of the tubes also had bubble-like structures ( Figure 3B ) , but were sometimes seen as blunt ( Figure 3C ) . Occasionally , the end of a tube was attached to a virion , with a constriction that suggested a detachment point of a nascent viral particle ( Figure 3C ) . Further ultrastructural information was obtained from tomographic reconstructions in which in silico 5 nm thick slices were extracted every 6 nm to yield a detailed examination of the 3-D volume comprising the tubes and their periphery ( Figure 4A–F ) . Tomographic analysis again revealed a dense core of the tube surrounded by a well-defined rim of lower density , similar to that described above for both tubes and virions ( Figure 3 ) . Further structural analysis of the low density rim of the tube revealed that its surface was marked by periodically arranged high electron densities ( 8–12 nm apart ) . The density of these dots appeared to follow a helical path around the dense core ( Fig . 4F ) . It is unclear whether the periodic structures represent viral or host proteins . The similarities between the ultrastructure of the tubular structures and progeny virions ( Figures 3 and 4 ) support the idea that the tubes may represent an assembly intermediate during virus production . SV40 , BKV and JCV viral DNA and proteins have been previously localized to PML-NBs by immunohistochemistry [17] , [28]–[30] , and we determined if PyV DNA and proteins were similarly localized . PML-NBs were identified using antibodies directed against the PML protein . Viral proteins were visualized by staining PyV-infected C57 mouse embryo fibroblasts ( MEFs ) for Tag at 22–28 hpi . Similar results were obtained when PyV-infected 3T3 fibroblasts were used ( data not shown ) . Using a fluorescently-labeled ( FISH ) probe specific for viral DNA , we found PyV DNA localized adjacent to PML-NBs in C57 MEFs at 24 hpi ( Figure 5A , top panel ) . Tag was localized diffusely throughout the nucleus , with a subset of staining in more intense puncta co-localizing with PyV DNA ( Figure 5A , bottom panel ) . At later times ( 28 hpi ) , the intensity of Tag staining in the nucleus increased while its localization pattern remained similar to that seen earlier during infection ( compare Figure 5A to Figure 5B ) . When co-stained for PML-NBs and Tag at 28 hpi , we found that Tag localized around the PML-NBs ( Figure 5B ) . At 28 hpi , PML-NBs appeared larger in size when compared to uninfected cells , consistent with previous observations for BKV and SV40 [30] , [32] . At either time , imaging through multiple z-planes revealed that the Tag staining surrounded the PML staining while the PyV DNA stained areas adjacent to PML . This pattern of localization is similar to that previously observed for SV40 Tag during SV40 infection [32] . PML-NBs have been associated with host proteins responsible for transcriptional regulation , apoptosis , and DNA damage repair . The DNA repair complex , MRN , is comprised of three interacting proteins , MRE11 , Rad50 and Nbs1 . MRN is recruited to sites of double-strand DNA ( dsDNA ) breaks , including those organized at or near PML-NBs [41] , [42] and MRE11 is recruited to SV40 sites of replication but later degraded [32] . Using a combination of immunohistochemistry and FISH , we determined whether MRE11 localized with PyV DNA during infection . In uninfected cells , the MRE11 staining was faint , presumably due to its diffuse localization throughout the nucleus ( data not shown ) . Upon infection , MRE11 staining was increased and became localized in discrete nuclear foci around the viral DNA ( Figure 5C , top panel ) . When co-stained for Tag and MRE11 , the MRE11 foci co-localized with the bright Tag puncta ( Figure 5C , bottom panel ) . These data indicate that PyV also may recruit MRE11 to sites of viral DNA replication . To identify protein components of the virus factory , thin sections of virus-infected cells were stained with antibodies against VP1 and PML , using immunogold electron microscopy . The anti-PML staining was observed throughout the nucleus and occasionally within the virion clusters and near the tubes , but was not specifically associated with the tubular structures ( Figure 6 ) . The tubular structures of the virus factory stained positively for VP1 ( Figure 6 ) . These data indicate that the tubular structures are comprised of VP1 , although the presence of other host proteins cannot be excluded . The PML protein has been suggested to function as a scaffold for other PML-NB-associated proteins . When PML expression is disrupted , the PML-NB-associated proteins ( i . e . , Sp100 , Daxx , SUMO-1 ) do not localize in nuclear dots when analyzed by immunofluorescence staining [43] . Since we observed that PyV DNA and Tag localized around PML-NBs , we determined their localization in MEFs isolated from PML-knockout mice ( PML−/− MEFs ) [44] . In PyV-infected PML−/− MEFs , PyV DNA was localized in small , punctate patches ( Figure 7A ) , similar to its localization in C57 MEFs and 3T3 fibroblasts ( Figure 5A; data not shown ) . In addition , Tag and MRE11 localization in infected PML−/− MEFs was similar to that observed in wild-type infected fibroblasts ( Figure 7B ) . Tubular structures were also observed in close proximity to virion clusters in these cells , similar to those seen in PyV-infected 3T3 cells and C57 MEFs ( Figure 6; data not shown ) . Immunogold labeling of PML−/− MEFs revealed that VP1 stained the virus factories similarly to the factories found in 3T3 cells ( Figure 6 ) . Together these data suggest that the PML protein is not necessary for the punctate localization of PyV DNA , its associated proteins , or the formation of virus factories . The observation that virus factories were present in PyV-infected PML−/− MEFs suggested that PyV replication is not impaired in these cells . To test whether PML−/− MEFs assembled and released virus , viral DNA ( Hirt DNA , [45] ) and supernatants from infected cells were isolated at various times after infection . We found that viral DNA accumulation was similar between 3T3 cells , C57 and PML−/− MEFs ( data not shown ) . At 3 and 4 days post-infection ( dpi ) , supernatants from cells were analyzed for the presence of packaged viral genomes . Supernatants were digested with DNase to remove unprotected DNA prior to proteinase K digestion , and then analyzed by qPCR for viral DNA . The supernatants from 3T3 cells , C57 and PML−/− MEFs contained similar amounts of virus particles by this assay ( Table 1 ) . Furthermore , these supernatants contained similar titers of infectious virus , as determined by plaque assay ( Table 1 ) . Therefore , PML is not required for PyV replication , assembly or release of infectious virus in vitro . The in vivo progression of PyV infection was compared between PML−/− mice [44] and the syngeneic mouse strain ( Sv129 ) using qPCR for viral DNA in isolated tissues . Sv129 or PML−/− mice were infected intraperitoneally with mouse polyomavirus strain A2 and mice were sacrificed 6 or 14 days later . Total DNA was isolated from homogenized tissue and PyV genomes were quantitated by qPCR ( Table 2 ) . The level of viral genome replication was similar between wild-type and PML−/− mice on days 4 and 6 ( Table 2; data not shown ) . On day 14 , however , although the kidneys had very similar PyV load in PML−/− mice and the wild type controls , the liver and lung tissues of PML−/− mice had lower viral genome levels than wild type mice . By day 14 , the adaptive immune response starts to control the infection , therefore data from mice infected for 6 days provided a more reliable readout of the ability of the virus to replicate in vivo .
Many viruses appear to assemble progeny virions at specific intracellular locations , often termed “virus factories . ” These sites may represent subcellular scaffolds where replicated genomes and capsid proteins spatially intersect in an efficient and coordinated manner to assemble virions . We used electron microscopy and tomographic 3-D reconstructions to identify possible sites and structural features for PyV virus assembly . We identified tubular structures in close physical association with progeny virions in the nucleus of PyV-infected mouse fibroblasts . The chronology of tube appearance suggests that these structures form at the initiation of virion assembly . Tubular structures were not observed in the cytoplasm of either infected or uninfected cells . We have termed the tubular structures , together with the associated virion clusters , virus factories . The tubular structures appear similar to “filamentous” structures previously observed in samples prepared by conventional , chemical fixation techniques and visualized by electron microscopy in PyV-infected baby hamster kidneys [46] and secondary mouse embryo cell cultures [47] , COS-7 cells transfected with expression plasmids encoding JCV viral proteins 29 , 48 , JCV-infected brain tissue [40] , [49] , [50] , and tissue samples from PML patients [51] , [52] . Consistent with the findings in these previous reports , we observed the accumulation of tubular structures early after infection and their relative absence late in infection when the nucleus was filled with virions [47] . In these early studies the tubular structures were also suggested to be assembly intermediates [46] , [47] , [53] , [54] . Tubular structures also have been reported in studies of papillomavirus isolated from human and rabbit warts [55] , and of PyV isolated from infected tissue culture cells [56] , [57] . Although we have not yet identified similar structures in BKV or SV40-infected cultured cells ( preliminary findings ) , the observation of tubular structures in histologic sections from PyV-infected mouse or hamster kidneys , PML patients , and human papillomavirus specimens supports their physiologic relevance to infection . Previous reports modeled two types of tubes from negatively-stained and low dose electron microscopic images . The tube types , termed hexamer or pentamer , are distinguished by the arrangement of VP1 pentamers within the tubes [56] , [57] . The hexamer tubes have an average diameter of 45–55 nm whereas the pentamer tubes have a diameter of ≈30 nm [55] , [56] . Thus , the tubular structures we observed during infection may represent the hexamer or pentamer tubes . We observed tubes with varying diameters ( 30–45 nm ) , which is consistent with the different diameters of the isolated tubes . The variation in diameter may be due to fixation and preservation of the samples isolated under various conditions . In contrast , however , the tubes we observed often had an electron dense core . In addition surrounding the electron-dense core ( Figure 4 ) we could identify a less dense rim associated with periodic densities ( “dots” ) spaced ≈8–12 nm apart . This distance is consistent with the distance between the centers of each pentamer in a hexamer tube [55]–[57] . Furthermore , the helical twist of pentameric subunits in pentamer/hexamer tubes is consistent with the apparent twisting of the filamentous structure around the tube core . Although the filamentous structures surrounding the tubes likely are comprised primarily of VP1 pentamers , it is possible that host proteins are also associated with the structures . The defining step in encapsidating large DNA viruses has been envisioned as the injection of the viral genome into a preformed capsid structure , similar to phage assembly . The current view of small DNA virus encapsidation is vague , hypothesizing that coat proteins are polymerized around the viral minichromosome in some manner . Our images suggest an alternative view . In comparison with virions in the same sections , the density surrounding the tubes corresponded to the density of the outer capsid of the virions , while the dense core was similar to the virion interior . Along with the occasional appearance of “budding” virions from the ends of the tubes , it might be envisioned that virion encapsidation proceeds in a manner akin to making sausage . The VP1 pentamers would first polymerize into tubes , and the viral chromatin then traverses the interior of the tubes until a terminal sealing event buds off the icosahedral particle . This model has an obvious relation to the assembly of large DNA viruses in that a preformed structure is subsequently filled with the viral genome . We present a model of icosahedral virus budding from the tubes in supporting information Figure S1 . Although PML-NBs have been previously speculated to be sites of viral DNA replication and possibly virion assembly , we found that the loss of the PML protein itself did not affect the punctate localization of PyV Tag or DNA ( Figure 7 ) , or the presence of tubular structures ( Figure 6 ) . Thus , while the nuclear bodies as defined by PML immunohistochemistry are not required for virus assembly , it is possible that other components of the PML-NBs may remain in a focal distribution to facilitate viral DNA replication , packaging and virion assembly . In particular the continued localization of MRE11 in puncta in the absence of PML ( Figure 7 ) suggests that important functions continue in discrete foci . As yet we have not been able to directly associate the virus factories we observe in the EM sections with replicating viral DNA , Tag , or PML seen by immunohistochemistry . Although the virion clusters appear in discrete patches and could represent the foci of viral DNA seen by FISH , the currently available antibodies for PML , MRE11 and Tag , have not performed well in the freeze-substituted sections for immunoEM , presumably because the epitopes were destroyed during sample preparation . However , the immunohistochemistry consistently localizes PyV DNA and Tag around the PML staining , suggesting that replication may be occurring adjacent to but not directly at the site of the factories . PML-NBs have been suggested to be part of an intrinsic cell defense against the replication of a variety of viruses [22] . Both PML and Sp100 are interferon-inducible genes [58] , [59] , and their protein levels correlate with the increased size and number of PML-NBs after IFN induction [60]–[63] . Large DNA viruses such as HSV-1 and Ad5 specifically target PML-NB structures during their replication cycle . PML-NBs are modified by the HSV-1 ICP0 protein , which induces proteasome degradation of the PML protein [64]–[66] whereas the Ad E4 ORF3 protein disrupts PML-NBs by reorganizing PML into fibrous strands [19] , [67]–[69] . Without these viral proteins , viral replication is reduced for both HSV-1 and Ad . Likewise , certain RNA viruses obtain a replication advantage when PML expression is disrupted . For example , when PML−/− mice were infected with either vesicular stomatitis virus ( VSV ) or lymphocytic choriomeningitis virus ( LCMV ) , an increase in virus production was observed [60] . However , PML-NBs do not serve as a site of assembly for these RNA viruses , and the anti-viral effects may be a consequence of a separate function of PML-NBs . Our findings represent the first study which demonstrates both in vitro and in vivo , that loss of PML expression does not affect virus replication . These results are consistent with data from other polyomaviruses ( BKV , JCV ) using cells with shRNA knockdown for PML or arsenite-treatment to disrupt PML protein expression [30] , [70] . Unlike JCV , however , we did not see an enhancement of virus production when PML was absent [70] . Furthermore , we did not see a rearrangement of PML-NBs such as observed during BKV infection [30] , although a slight increase in the size of PML-NBs during infection was noted . The structural features of virion assembly factories are beginning to be defined for many viruses . Viruses frequently yield important insights for cell biology , and the structures facilitating virus assembly may also inform hypotheses concerning the organization of important nuclear functions such as for DNA repair . The multiple functions assigned to PML-NBs are undergoing continued revision with respect to host-virus interactions . The role of PML-NBs in PyV replication is complex , and PML-NB associated proteins other than PML itself may provide the necessary architectural foundation for the tubular structures that appear to be the PyV factories .
This study was carried out in strict accordance with the recommendations in the Guide for the Care of Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of the University of Massachusetts Medical School ( Approval number: 305 ) , following recommendations by the Office of Laboratory Animal Welfare ( OLAW ) . All mouse strains were maintained in specific pathogen–free conditions in the animal facilities of the University of Massachusetts Medical School . All efforts were made to minimize suffering and provide humane treatment to the animals included in the study . A31 3T3 fibroblasts were grown in DMEM ( D6429; Sigma ) supplemented with 10% bovine calf serum ( BCS ) and penicillin/streptomycin . C57 mouse embryonic fibroblasts ( MEF ) were obtained from ATCC ( SCRC-1008; Manassas , VA ) and served as a wild-type MEF control . PML−/− MEFs were a gift of Gerd Maul [43] , [44] . Both MEFs were grown in DMEM supplemented with 20% fetal bovine calf serum ( FBS ) and antibiotics . Virus strain NG59RA [71] was used for all cell culture infections and strain A2 [72] for mouse infections . Cell culture infections were carried out as described previously [73] . Cells were infected at a multiplicity of infection of 10–20 pfu/cell for 32–34 hours for electron microscopy and replication studies , and 30–40 pfu/cell for 22–24 hours for immunofluorescence analysis . A higher MOI was used for immunofluorescence studies to compensate for cell loss during the staining process . Similar results were observed when an MOI of 10–20 pfu/cell was used although it was more difficult to identify infected cells due to loss during staining . Similar results were observed at 32–34 hours post-infection ( hpi ) when cells were stained for immunofluorescence analysis . However , cell loss was significantly increased at later times and the fluorescence signal was saturated . Thus , to define protein localization , we chose to show earlier times after infection for the immunofluorescence studies . T-antigen was detected using rat anti-T-antigen ( E3 , gift of Tom Benjamin ) . PML was detected using mouse anti-PML ( ALX-804-816; Enzo , Plymouth Meeting , PA ) and MRE11 was detected using rabbit anti-MRE11a ( B1447; LSBio , ) . Primary antibody dilutions for rat anti-Tag , mouse anti-PML , and rabbit anti-MRE11a were 1∶5000 , 1∶1000 and 1∶1000 , respectively . Secondary antibodies were either AlexaFluor 488- or AlexaFluor 594-conjugated ( Invitrogen ) and were diluted 1∶1000 . All antibodies were diluted in 10% FBS/PBS . Cells were grown on acid-etched , poly-L-lysine-coated coverslips and infected with NG59RA as described above except cells were harvested at 22–24 hpi . Cells were prepared for immunofluorescence staining according to Zhao et al [32] , with some modifications . Briefly , cells were washed 3 times with cold phosphate-buffered saline ( PBS ) solution followed by cytoskeleton buffer ( CSB ) [10 mM piperazine-N , N-bis ( 2-ethanesulfonic acid ) ( PIPES ) , pH 6 . 8 , 100 mM NaCl , 300 mM sucrose , 1 mM MgCl2 , 1 mM EGTA] . Soluble proteins were pre-extracted for 3 min with cold CSB containing 0 . 5% Triton X-100 and protease inhibitors ( Complete mini tablets , Pierce ) at 4°C . The cells were then washed with PBS , fixed in 4% paraformaldehyde ( PFA ) in PBS for 30 min , washed with PBS , and blocked with 10% FBS/PBS at 4°C . Samples were incubated with Image-iT FX signal enhancer ( I3693; Invitrogen ) for 20 min followed by incubation with primary antibodies for 1 hr at room temperature , washed with PBS , and incubated with secondary antibodies for 1 hr at 22°C . Stained cells were mounted onto glass slides with ProLong anti-fade reagent containing Dapi ( P36935; Invitrogen ) and allowed to incubate at 22°C overnight . To detect Py viral genomes after infection , we used nick-translation to label DNA probes specific for PyV DNA . Briefly , the entire Py viral genome ( NG59RA ) was cloned into pUC18 at BamHI ( pUC-PyV ) and 2 µg plasmid DNA was labeled with SpectrumRed using the Vysis Nick Translation Kit ( Abbott Molecular , Des Plaines , IL ) , according to the manufacturer protocol . Labeled DNA was ethanol-precipitated with herring sperm and human Cot-1 DNA and resuspended in 20 µl cDen-Hyb ( Insitus Biotechnologies , Albuquerque , NM ) to yield a final probe concentration of 100 ng/µl . FISH analysis was performed as described previously [28] , with some modifications . Briefly , cells grown on coverslips were infected , fixed and immunostained for viral or host proteins , as described above . Immunostained cells were fixed a second time with 4% PFA in PBS to crosslink bound antibodies followed by treatment with 0 . 2 mg/ml RNase Type III ( Sigma ) in 2X SSC at 37°C for 15 min and washed in 2X SSC 3 times . The PyV DNA probe was diluted in cDenHyb and hybridized to samples for 2 min each at 80°C , 70°C , 60°C , 50°C and 45°C followed by an overnight incubation at 37°C . Coverslips were washed at 45°C with 1 . 5X SSC , 50% formamide/1 . 5X SSC , and 1 . 5X SSC for 5 min each . Stained cells were mounted onto glass slides with ProLong anti-fade reagent containing Dapi ( Invitrogen ) and allowed to incubate at 22°C overnight . Imaging of fixed cells was performed with an inverted fluorescence microscope ( TE2000-U; Nikon ) equipped with an electron-multiplying charge-coupled device camera ( Cascade II; Photometrics ) and a Yokogawa spinning disc confocal system ( CSU-Xm2; Nikon ) . Images were taken with a 60X NA 1 . 4 oil objective using MetaMorph ( version 7 . 0; MDS Analytical Technologies ) software . The data from each channel ( excitation wavelengths at 488 , 543 and 633 nm ) were collected sequentially using the appropriate band-pass filters built into the instrument . For z-stacks , optical slices were taken at 0 . 2 µm increments . Data sets were processed using ImageJ ( National Institutes of Health ) software . Data are shown as an extracted slice from the z-stacks . Infected cells were trypsinized and collected by centrifugation . The cell pellet was resuspended in growth media supplemented with either 20% dextran ( 40 , 210 Da , Sigma–Aldrich ) or 150 mM mannitol ( M9647; Sigma ) . For dextran-prepared samples , 3 µl of the cell suspension was deposited into an aluminum planchette ( Engineering Office M . Wohlwend GmbH , Sennwald , Switzerland ) with a sample depth of 100 µm and vitrified in the BalTec HPM 010 under a pressure of 2000 bar by a jet of liquid nitrogen applied on the carrier . For mannitol-prepared samples , cells were collected by centrifugation , the supernatant was poured off and a slurry was made by resuspending the cell pellet in the remaining supernatant . 1–5 µl of the cell slurry were transferred to an aluminum planchette and frozen as described above . The frozen cell suspensions were stored in liquid nitrogen prior to cryo-substitution and plastic embedding . Dextran-prepared samples: High-pressure frozen cells were cryo-substituted with 1% osmium tetroxide/0 . 2% uranyl acetate/acetone over three days at −90°C with a gradual warming to room temperature over a period of 2 days . Samples were removed from planchettes , and rinsed with acetone several times to remove residual osmium and uranyl acetate . Cells were infiltrated with increasing concentrations of Epon-Araldite 802 epoxy resin ( Electron Microscopy Sciences , Port Washington , PA ) over a period of 2 days , with 3 final incubations of 100% Epon to remove residual acetone . Samples were polymerized in BEEM embedding capsules ( Electron Microscopy Sciences ) by addition of DMP-30 accelerator and incubation at 60°C for 2 days . Mannitol-prepared samples: High-pressure frozen cells were cryo-substituted and prepared as described above for Epon epoxy resin embedding . For samples to be used for immunogold-labeling , frozen cells were cryosubstituted with 0 . 1% uranyl acetate/0 . 25% glutaraldehyde/acetone at −75°C for 3 days followed by a gradual warming to −35°C over 12 hr . Cells were removed from the planchettes and rinsed with −35°C acetone and infiltrated over 3 days with increasing concentrations of Lowicryl/HM20 resin ( Electron Microscopy Sciences ) , with 3 final changes of 100% HM20 . Resin-infiltrated cells were polymerized in BEEM capsules at −35°C under ultraviolet light for 2 days . Samples and resin were maintained at −35°C during infiltration steps in an automated freezing system ( AFS; Leica Microsystems ) . Epon-embedded sections were cut to a thickness of 70 nm with an UltraCut–UCT microtome ( Leica Microsystems ) using a diamond knife ( Diatome , Biel , Switzerland ) . Sections were picked up on Formvar-coated copper slot grids and stained with 2% aqueous uranyl acetate for 10 min , followed by Reynold's lead citrate [74] for 4 min . Images were obtained on a Philips CM100 electron microscope operating at 80 kV . For immunogold labeling , Lowicryl-embedded samples were sectioned at a thickness of 45 nm ( anti-PML ) or 70 nm ( anti-VP1 ) and picked up on Formvar-coated , copper slot grids . Sections were fixed with 0 . 5% paraformaldehyde ( diluted from a fresh stock ( 32% , Electron Microscopy Sciences ) in PBS ) for 15 min at 22°C . The grids were rinsed with PBS and blocked in 1% milk/PBST* ( PBS/0 . 02% Tween 20 ) in a humidified chamber for 30 min at room temperature followed by incubation in the same chamber on droplets of primary antibody diluted in 1% milk/PBST* for 2 hr at 22°C . The grids were washed in a stream of PBS , blotted to remove excess PBS and incubated with gold-conjugated secondary antibody for 1 hr at 22°C . Samples were washed in a stream of PBS followed by a distilled water wash , air-dried and post-stained as described above except incubation times were reduced to 4 min and 2 min for methanolic uranyl acetate and lead citrate , respectively . Samples were imaged as described above . Primary antibody dilutions for rabbit anti-VP1 ( I58 ) and rabbit anti-PML ( ab67761; AbCam , Cambridge , MA ) were 1∶2000 and 1∶80 , respectively . Secondary antibodies were 10 or 15 nm gold-conjugated goat anti-rabbit IgG ( Ted Pella Inc . , Redding , CA ) and were diluted 1∶20 . All antibodies were diluted in 1% milk/PBST* . Semi-thick Epon-embedded sections ( ∼300 nm ) were stained with 2% uranyl acetate as described above for thin Epon-embedded sections . Colloidal gold particles ( 15 nm; BBI Research , Inc . , Madison , WI ) were place on both grid surfaces to serve as fiducial markers for subsequent image alignment . Sections were imaged on a FEI Tecnai F20 microscope ( FEI Company Ltd . , Eindhoven , The Netherlands ) operating at 200 kV and images were collected on a 4K by 4K CCD Ultrascan camera ( Gatan , Inc . , Pleasanton , CA ) . Dual-axis tilt series data sets were acquired using the SerialEM software package [75] . 1×1 , 2×1 or 2×2 montages were acquired on either a single section or up to six serial sections ( Figure 1D and Supporting information Video S1 ) . The nominal defocus was set to 0 . 5 µm and the pixel size of the data varied between 0 . 764 and 1 . 206 depending on the magnification used . Tomographic reconstructions were produced by the IMOD software package [76] . 3-D structures of interest were surface rendered using tools available in the IMOD software package ( Figures 1D and 2C and videos ) . 105 cells per 60mm dish were seeded 12–16 hrs before infection with NG59RA . At this density , the cells were 50–60% confluent at the time of infection . Virus preparation and infection was carried out as described above . At times indicated , supernatants were transferred to 15 ml conical tubes and saved . Cells remaining on the plate were treated with neuraminidase ( NA ) Type V ( Sigma ) diluted in NA buffer ( 10 mM Hepes , pH 5 . 6/1 mM CaCl2/1 mM MgCl2/5 mM KCl ) at 37°C for 30 min . The NA supernatant was collected and combined with the cell supernatant . The plates were washed with PBS 3 times and each wash was collected and combined with supernatants . The combined supernatants and washes were stored at −20°C and are referred to as the “viral supernatant . ” Viral DNA was isolated from viral supernatants using the method described for Ad supernatants [77] with some modifications . Briefly , 50 µl viral supernatant was digested with DNase ( RQ1 , Promega ) at 37°C for 1 hr . DNase-treated samples were incubated with Proteinase K ( Fermentas ) for 37°C for 1 hr . The viral DNA was purified using the Wizard DNA Clean-up System ( Promega ) according to the manufacturer protocol except 80% ethanol was used in place of isopropanol during the wash . The viral DNA was eluted with 50 µl prewarmed ( 80°C ) milli-Q water and stored at −20°C until ready for use . Primer Express 3 . 0 software ( Applied Biosystems , Warrington , UK ) was used to design probes and primers for amplification of a 67-bp region of the mouse polyomavirus genome ( NCBI accession # NC_001515 ) . PCR primers were synthesized by IDT and the TaqMan probe was synthesized by Applied Biosystems . The following primers were used: PyV VP1 forward primer , 5′TGGGAGGCAGTCTCAGTGAAA3′; PyV VP1 reverse primer , 5′TGAACCCATGCACATCTAACAGT3′; PyV VP1 probe , 5′CCGAGGTGGTGGGCTCTGGC3′ . The TaqMan probe was labeled with FAM ( 6-carboxy-fluorescein ) at the 5′ end and a quencher , TAMRA ( 6-carboxy-tetramethylrodamine ) at the 3′ end . Optimal concentrations of probe and primer were determined using a concentration matrix , as described in Applied Biosystems protocols . Quantitative PCR ( qPCR ) reactions were prepared in 96-well optical plates ( Applied Biosystems ) in a volume of 25 µl . Each reaction contained 200 nM TaqMan probe , 900 nM of each forward and reverse primer , 12 . 5 µl TaqMan Master Mix ( Applied Biosystems ) and 5 µl purified viral DNA or DNA standards . The intensity of fluorescence of the reporter label was normalized to the ROX Passive Reference , supplied in the master mix solution . DNA amplification was carried out using an Applied Biosystems 7500 Real-Time PCR Sequence Detection System using cycling conditions of 50°C for 2 min , 95°C for 10 min followed by 40 cycles of 95°C for 15 sec , 60°C for 1 min . For each run , duplicates of seven dilutions of the viral standard DNA ( from 0 . 1 ng to 5×10−5 ng; pGEX-VP1 plasmid DNA ) , viral DNA samples and no template controls were simultaneously subjected to analysis . The number of genomes was determined as previously described [77] and is reported as “nuclease-resistant genomes ( NRG ) per ml . ” NIH/3T3 cells were resuspended in absorption medium ( DMEM/0 . 1% BSA ) at a density of 1×106 cells per ml . The virus stock was serially-diluted in absorption medium . Equal volumes of the cell suspension and viral dilution were mixed at 37°C , 5% CO2 for 1 . 5 hrs with gentle agitation every 15 minutes . Complete growth media ( DMEM/5% BCS/penicillin/streptomycin ) was added to the infected cell suspension and transferred to a six-well plate in triplicate ( 1×105 cells/well ) . Cells were incubated at 37°C for 5 to 6 hrs to allow for cell adherence after which the growth media was gently removed and pre-warmed ( 42°C ) overlay medium ( DMEM/5% BCS/1 µM dexamethasone/penicillin/streptomycin/0 . 4% agarose ) was added to each well and allowed to solidify . Samples were incubated at 37°C and 5% CO2 for 6–8 days . After incubation , the overlay agar was carefully removed and each well was washed with cold PBS . The monolayer was stained with crystal violet solution ( 1% crystal violet/formaldehyde ) for 8 min , the stain decanted and each well washed with cold PBS . After PBS washes , the plates were inverted and allowed to dry before counting the plaques . PML−/− mice were obtained from the NCI and 129S1/Svlmj wild-type control mice were purchased from Jackson Laboratories . Mice were infected intraperitoneally at 2 months of age with 5×105 pfu per mouse with PyV strain A2 . Mice were sacrificed 6 and 14 days post-infection and their organs ( spleen , kidney , liver , lung ) were isolated to measure viral load by qPCR . To determine the viral load in infected mouse tissues by qPCR , DNA was prepared from organ homogenates by proteinase K ( Sigma ) digestion at 55°C overnight , followed by phenol extraction and RNase-A treatment ( 10 units/µl , Promega ) . The PCR amplification was performed as described previously [78] . Briefly , a 50 µl reaction mixture with 2X SYBR green master mix ( 4309155 , Applied Biosystems ) and 0 . 1 mM each of forward and reverse primers ( Invitrogen ) was prepared . The following primers were used: β actin forward , 5′CGA GGC CCA GAG CAA GAG AG3′; β actin reverse , 5′CGG TTG GCC TTA GGG TTC AG3′; PyV VP1 forward , 5′CCC CCG GTA CAG GTT CAG TCC CAT CAT3′; VP1 reverse , 5′GGC ACA ACA GCT CCA CCC GTC CTG CAG3′ . The amplification for VP1 started with one cycle at 95°C for 10 min , 37 cycles of 95°C for 30 sec , 65°C for 20 sec , 72°C for 45 sec . PCR amplification with the β actin primers started with one cycle at 95°C for 10 min , then 40 cycles of 95°C for 30 sec , 62°C for 25 sec , and 72°C for 25 sec . Negative controls included a sample with no DNA template and DNA from uninfected mouse organs . Three-fold serial dilutions of DNA prepared from uninfected mouse organs ( spanning 1 µg–31 ng ) were used to generate a standard curve for β actin PCR . For PyV we used a recombinant plasmid containing the VP1 coding sequences and made dilution series from 2×108 copies to 20 copies . All the reactions were run in duplicates . The obtained PyV copy numbers were normalized for β actin which reflected the amount of mouse genomic DNA present , and the results were expressed as PyV genome copies per µg organ DNA . | Polyomaviruses are infectious pathogens of mammals and birds that have been linked to the development of cancers in their hosts . Members of the polyomavirus family are associated with human disease , such as JCV and BKV , and over the past few years , several more human polyomaviruses ( WUV , KIV and MCV ) have been discovered in immune-suppressed individuals . We are studying the way in which these viruses assemble in cells in order to identify critical points where anti-viral therapies could target these viruses . Using a structural , biochemical and cell biological approach , we set out to define sites of virus assembly and virus intermediates . We identified virus-specific structures that we termed “virus factories” . We believe that these sites serve as an assembly line for the production of new viruses . Our study provides new evidence for the presence and composition of virus assembly factories , and identifies a host protein that may be important for infection by polyomaviruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology",
"biology",
"microbiology",
"viral",
"replication",
"viral",
"structure"
] | 2012 | Virion Assembly Factories in the Nucleus of Polyomavirus-Infected Cells |
Genome duplications increase genetic diversity and may facilitate the evolution of gene subfunctions . Little attention , however , has focused on the evolutionary impact of lineage-specific gene loss . Here , we show that identifying lineage-specific gene loss after genome duplication is important for understanding the evolution of gene subfunctions in surviving paralogs and for improving functional connectivity among human and model organism genomes . We examine the general principles of gene loss following duplication , coupled with expression analysis of the retinaldehyde dehydrogenase Aldh1a gene family during retinoic acid signaling in eye development as a case study . Humans have three ALDH1A genes , but teleosts have just one or two . We used comparative genomics and conserved syntenies to identify loss of ohnologs ( paralogs derived from genome duplication ) and to clarify uncertain phylogenies . Analysis showed that Aldh1a1 and Aldh1a2 form a clade that is sister to Aldh1a3-related genes . Genome comparisons showed secondarily loss of aldh1a1 in teleosts , revealing that Aldh1a1 is not a tetrapod innovation and that aldh1a3 was recently lost in medaka , making it the first known vertebrate with a single aldh1a gene . Interestingly , results revealed asymmetric distribution of surviving ohnologs between co-orthologous teleost chromosome segments , suggesting that local genome architecture can influence ohnolog survival . We propose a model that reconstructs the chromosomal history of the Aldh1a family in the ancestral vertebrate genome , coupled with the evolution of gene functions in surviving Aldh1a ohnologs after R1 , R2 , and R3 genome duplications . Results provide evidence for early subfunctionalization and late subfunction-partitioning and suggest a mechanistic model based on altered regulation leading to heterochronic gene expression to explain the acquisition or modification of subfunctions by surviving ohnologs that preserve unaltered ancestral developmental programs in the face of gene loss .
Understanding the evolution of gene functions during vertebrate evolution is important for the proper interpretation of comparative analyses , especially when using model organisms to understand human gene functions . Gene duplication has been proposed to facilitate the evolution of gene functions [1] , and the mechanisms of neofunctionalization and subfunctionalization may play a role [1]–[3] ( reviewed in [4] ) . Human gene families show the signatures of two rounds of whole genome duplication ( R1 and R2 ) that occurred during early vertebrate evolution [1] , [5]–[14] ( but see [15] ) . Mutations in gene copies that arose in these R1 and R2 events often cause related diseases ( for example , osteogenesis imperfecta ( COL1A1 ) and spondyloepiphyseal dysplasia ( COL2A1 ) , bullous erythroderma ichthyosiformis ( KRT1 ) and epidermolysis bullosa ( KRT5 ) , and syndactyly type II ( HOXD13 ) and hand-foot-uterus syndrome ( HOXA13 ) ) . Comparative analysis shows that fish genomes have two co-orthologs for many human genes as a result of a third round of genome duplication ( R3 ) that occurred at the base of the teleost radiation [16]–[28] . Early on , S . Ohno [1] recognized the relevance of increased genetic diversity after genome duplication , and in his honor , gene duplicates originated by genome duplication are called “ohnologs” [29] . This term is useful because of the special properties that ohnologs possess at their birth compared to duplications that arise by other mechanisms such as unequal crossing-over , tandem gene duplication , or retrotransposition . While many studies focus on how gene duplications can facilitate the acquisition of evolutionary innovations during vertebrate evolution , less attention has been focused on the evolutionary impact of lineage-specific gene losses . Differential ohnolog loss is important because it decreases genetic diversity within a species but increases genetic diversity between species . Loss of one copy of a pair of fully redundant gene duplicates should not usually have significant impact , but duplicate loss after functional divergence can have evolutionary consequences . Reciprocal paralog loss in different lineages can affect a species' biology , decrease evolvability , and diminish adaptability to changing environments [30]–[34] . In other cases , gene loss can be adaptive , and thus relevant for a species' evolution , perhaps even for human origins [35] . Furthermore , reciprocal loss of even fully redundant gene duplicates in two populations may contribute to speciation [33] , [36] . Global estimations of gene loss in fully sequenced vertebrate genomes have been inferred by massive phylogenetic reconstructions of gene families [37]–[39] . Large-scale analyses , however , are sensitive to uncertainties of phylogenetic analysis , for example , asymmetric rates of evolution among paralogs can affect tree topologies and generate gene phylogenies that are not congruent with the species phylogenies of which they are a part [40] , [41] . Furthermore , published genome-wide studies have not addressed gene function . In principle , gene functions that are associated exclusively with a certain gene may disappear if the gene is lost . It is possible , however , that exclusive gene functions might not disappear in situations in which a surviving paralog might acquire or maintain the expression domain of the lost paralog , and thereby the ancestral developmental or physiological program can remain unaltered [42] , [43] . Because the evidence for gene loss is negative and can pass unnoticed and is subject to uncertainties in the completion or assembly of sequenced genomes and in copy number polymorphisms [44] , [45] , the impact of gene loss in the evolution of function of surviving paralogs is under-investigated . Identification of gene loss is especially important to avoid misinterpretations when human gene functions are inferred from the study of model organisms that might have suffered lineage-specific paralog loss , so that the model has no true ortholog of the phylogenetically most closely related human gene , or vice versa . To evaluate the evolutionary relevance of gene loss on the functions of surviving paralogs , it is first important to understand gene phylogeny . For genes lost following large-scale genome duplications , conserved syntenies can identify duplicated genomic regions and provide evidence for gene loss , often even in situations lacking a proper outgroup [46] . Genes lost after genome duplication events have been called “ohnologs gone missing” ( ogm ) , and their identification is important to properly distinguish orthologs from other types of paralogs [32] , [47] , [48] . We propose that identification of gene loss by automated comparative genomic analysis of conserved syntenies can: ( 1 ) help resolve uncertain gene phylogenies; ( 2 ) help discriminate cases of evolutionary innovations from evolutionary simplifications; ( 3 ) facilitate understanding of the diversification of gene functions among species; and , importantly , ( 4 ) improve functional connectivity of human and model organism genomes . To explore the roles of gene loss in a functional context , work reported here focuses on the vertebrate Aldh1a retinaldehyde dehydrogenase gene family ( formerly known as Raldh ) as a case study . Understanding the evolution of Aldh1a genes is important because this family encodes enzymes responsible for the synthesis of retinoic acid ( RA ) , the active derivative of vitamin A ( retinol ) . In humans , as in other vertebrates , RA plays important roles during embryogenesis , for example , in axial patterning , limb development , and differentiation of eyes and nervous system , as well as functioning in adult organ homeostasis ( recently reviewed in [49] , [50] ) . Alterations of RA metabolism can lead to human pathologies including breast and prostate cancers , osteoporosis , rheumatoid arthritis , dermatologic diseases , developmental anomalies and premature births . The evolutionary origin of the Aldh1a family probably predates the origin of stem bilaterians [51] , [52] , but the ability of the Aldh1a enzyme of basally diverging bilaterians to synthesize RA remains unknown . Aldh1a likely arose by duplication of an ancestral gene related to the Aldh2 gene family , which encodes a mitochondrial Aldh that plays a major role in acetaldehyde oxidation and is broadly represented in most extant organisms from bacteria to humans [53] . Humans and many other vertebrates have three genes that encode Aldh1a family enzymes: ALDH1A1 , ALDH1A2 and ALDH1A3 [54] . Studies of model organisms such as mouse , chicken , frog and zebrafish have provided insights into the roles of each Aldh1a gene in the synthesis of RA ( reviewed in [49] , [50] , [55]–[58] ) . Variation in Aldh1a gene number in different animal lineages has been hypothesized to be relevant to animal evolution due to potential effects of RA metabolism on the mechanisms of development [59]–[61]; reviewed in [62] . Rodents have a fourth Aldh1a paralog that is mostly expressed in kidney ( termed , Aldh1a4 in rat [63] , and its ortholog Aldh1a7 in mouse [64] ) ; these genes originated by a tandem gene duplication in the rodent lineage after it diverged from the human lineage . Experiments using a heterologous Xenopus system to express mouse Aldh1a7 suggested that Aldh1a7 might not be involved in RA synthesis [64] . In contrast to rodents with four Aldh1 genes , most teleost fish have just two , aldh1a2 and aldh1a3 , but they lack aldh1a1 [59] , [65] . Phylogenetic relationships of vertebrate Aldh1a1 genes are still controversial , and whether Aldh1a1 is a tetrapod innovation or its absence from teleosts is due to gene loss is still unknown . Furthermore , the functional consequences of these gene copy number variations have not yet been investigated . Here , we show how comprehensive comparative genomic analyses of syntenic conservation provides a framework necessary for the examination of the general mechanisms by which lineage-specific gene loss can impact the functions of surviving paralogs . This work reveals multiple losses of Aldh1a ohnologs and proposes an evolutionary genomic model that reconstructs the history of Aldh1a-related vertebrate chromosomes and the evolution of Aldh1a gene functions during and subsequent to the R1 , R2 , and R3 genome duplications . Results show that acquisition or modification of expression domains by surviving paralogs may lead to lineage-specific innovations that preserve unaltered ancestral developmental programs in the face of gene loss . This work highlights the importance of comparative genomics for understanding the historical basis of gene loss , and to improve functional connectivity between model organism and human genomes .
To understand the history of gene gain and loss in the Aldh1a family , it is important to first understand the phylogeny of family members . Unfortunately , evolutionary relationships among vertebrate Aldh1a paralogs are currently unclear . In one analysis , the three vertebrate Aldh1a clades collapsed to an unresolved trichotomy [59] , and in another , Aldh1a2 and Aldh1a3 appeared as sister groups ( Aldh1a1 , ( Aldh1a2 , Aldh1a3 ) ) , supported by low bootstrap values [65] . These problems may stem from sequence similarities among the Aldh1a1 , Aldh1a2 and Aldh1a3 proteins and the use of the evolutionarily distant mitochondrial Aldh2 family to root the tree . To overcome this uncertainty , we turned to a chordate outgroup , the cephalochordate amphioxus , whose lineage diverged from that of the vertebrates before the R1 and R2 events [66] , [67] . Amphioxus has both Aldh1a and Aldh2 gene families [59] , and hence its Aldh1a genes are much more closely related to vertebrate Aldh1a1 genes than is the Aldh2 gene family . We found that several different phylogenetic methodologies , including Bayesian inference , Maximum-likelihood , gamma-corrected Neighbor-Joining and Maximum-Parsimony all agreed on the same tree topology ( ( Aldh1a1 , Aldh1a2 ) , Aldh1a3 ) ) , with Aldh1a1 and Aldh1a2 as sister groups ( Figures 1 and S1 ) . This phylogeny differs from both published results: the trichotomy result and the view of Aldh1a2 and Aldh1a3 as sister clades [59] , [65] . Our results still provided only a moderately high probability of 0 . 76 supporting the Aldh1a1/2 clade under the Bayesian phylogenetic inference ( Figure 1 ) ; thus , phylogenetic analysis alone is insufficient to definitively resolve Aldh1a relationships . To further test historical relationships among Aldh1a paralogs , we examined a data set independent of Aldh1a gene sequence by conducting comparative genomic analyses of the entire genomic neighborhoods ( GN ) surrounding Aldh1a genes in the genomes of humans and other vertebrates . The results of our phylogenetic analysis ( ( Aldh1a1 , Aldh1a2 ) , Aldh1a3 ) ( Figure 1 ) implies that the duplication event that gave rise to Aldh1a1 and Aldh1a2 was more recent than the duplication event that gave rise to Aldh1a3 and the ancestral Aldh1a1/2 gene . If the duplication events that produced the Aldh1a family involved whole genomes or large chromosomal segments , then the phylogenic hypothesis of relationships ( Figure 1 ) predicts more syntenic conservation between the genomic neighborhoods ( GN ) surrounding Aldh1a1 and Aldh1a2 than between the genomic neighborhood of Aldh1a3 and either Aldh1a1 or Aldh1a2 . To test this hypothesis , we conducted a comparative genomic analysis of conserved synteny among the genomic neighborhoods of each ALDH1A paralog in the human genome . The three human ALDH1A genes are located on two chromosomes: ALDH1A1 is on Hsa9 ( human chromosome 9 ) , while ALDH1A2 and ALDH1A3 are on Hsa15 separated by 43 megabases ( Mb ) . We first made a composite dotplot to represent the genome-wide distribution of the paralogs of all genes within a 10 Mb-window surrounding each human ALDH1A gene throughout the 23 human chromosomes ( y-axis ) ( we refer to this set of genes as ALDH1A-neighbor paralogs ( red , blue and green crosses in Figure 2A ) ) . Table S1 lists gene names , reference numbers , genomic positions and outgroup ( i . e . Branchiostoma floridae and Ciona intestinalis ) gene information used to construct each paralogy group in the dotplot . This plot showed that while some ALDH1A-neighbor paralogs appeared randomly scattered throughout the genome , some chromosomal regions contained a concentration of ALDH1A-neighbor paralogs ( yellow and pink boxes in Figure 2A ) . These chromosome regions with syntenic conservation to ALDH1A-neighbor paralogs likely represent chromosome fragments that were duplicated during the whole genome duplication events R1 and R2 and are historically related to the expansion of the Aldh1a family . The presence of ALDH1A-neighbor genes conserved among ALDH1A genomic neighborhoods ( pink-shaded dotted boxes ) suggests that the ALDH1A family expanded by large-scale genome duplications rather than by local tandem gene duplications . The dotplot analysis also identified four genomic regions that share syntenic conservation with ALDH1A genomic neighborhoods , but do not contain ALDH1A genes ( yellow-shaded boxes on Hsa1 , Hsa5 , Hsa9 and Hsa19 ) . The paralogs of each gene contained in these four yellow boxes were also included in the dotplot ( Figure 2A: golden , black , pink and brown crosses ) . In principle , the existence of the yellow-boxed regions that lack ALDH1A paralogs but show syntenic conservation with the ALDH1A genomic neighborhood could be explained by genome duplications followed by a loss of the ALDH1A paralog ( i . e . ALDH1A ohnologs gone missing ) , or alternatively by the translocation of a portion of the genomic neighborhood away from the ALDH1A gene itself . When gene functions are compared among different organisms , it is important to distinguish whether the compared genes are orthologs or paralogs . In some cases , reciprocal loss of paralogs in different organisms can lead to the misinterpretation of paralogs as orthologs . Intriguingly , while in most vertebrates Aldh1a2 and Aldh1a3 are on the same chromosome separated by an intervening region of few tens of megabases , in rodents Aldh1a2 and Aldh1a3 are on different chromosomes [63] , [64] . In rats , for instance , Aldh1a3 is on the same chromosome as Aldh1a1 rather than being on the same chromosome as Aldh1a2 as in human . This arrangement would be expected if the rat Aldh1a3 gene were a paralog rather than an ortholog of human ALDH1A3 . Phylogenetic analysis provided strong support for the conclusion that all vertebrate Aldh1a3 genes are orthologs [59] , [65] , [69] , but evidence for an ALDH1A3 ohnolog gone missing from the human genome raises the possibility of reciprocal paralog loss that would have caused human and rodent Aldh1a3 genes to be paralogs rather than orthologs . To see whether the mouse Aldh1a3 gene is orthologous to human ALDH1A3 or to ALDH1A3-ogm , we first constructed a dotplot that displayed the distribution of the mouse orthologs of human genes within 10 Mb of ALDH1A1 , ALDH1A2 , ALDH1A3 and ALDH1A3-ogm ( Figure 3A and Table S2 ) . The dotplot revealed that most mouse orthologs of the human ALDH1A-neighbor genes tightly clustered on four mouse chromosomes ( Mmu7 , Mmu9 , Mmu13 , and Mmu19 ) ( Figure 3A ) . Next , we compared these four mouse chromosomes to their orthologons on human chromosomes Hsa5 , Hsa9 and Hsa15 in a circleplot ( Figure 3B ) . These analyses identified four clusters of orthology in the Synteny database [48] that unequivocally related mouse orthologs of human ALDH1A1 , ALDH1A2 , ALDH1A3 , and ALDH1A3-ogm genome neighborhoods ( Figure S2 ) . The identification of a genomic region on Mmu13 that lacks any Aldh1a gene but that nevertheless possesses orthologous syntenic conservation to ALDH1A3-ogm genomic neighborhood on Hsa5 ( golden bundle in Figure 3B ) provides strong evidence that the loss of Aldh1a3-ogm predated the split between the lineages leading to humans and rodents , and discards the hypothesis of reciprocal paralog loss . These results conclusively rule out the hypothesis that the rodent Aldh1a3 is an ortholog of human ALDH1A3-ogm , and independently supports orthologous relationships between human and mouse Aldh1a3 genes inferred by phylogenetic methods ( Figure 1 ) . Because the number of Aldh1a paralogs detected in genome databases is lower in teleosts than in tetrapods [59] , [65] , we performed a comparative genomic analysis of conserved synteny between Aldh1a genomic neighborhoods in the genomes of three teleosts and human to learn the historical basis of different numbers of gene family members ( Figure 4 ) .
Retinoic acid plays important morphogenetic roles in chordate embryonic development . The recent identification of components of the RA genetic machinery in non-chordate deuterostomes and in protostomes opens the possibility that expansion and reduction in RA-related gene families could have played a role in the developmental diversification of bilaterians [51] , [52] . The Aldh1a gene family , which encodes enzymes that synthesize RA , has expanded independently several times during the evolution of the three chordate subphyla , the Cephalochordata , Urochordata and Vertebrata [59] . Within vertebrates , the expansion of the Aldh1a family generated three main paralogs - Aldh1a1 , Aldh1a2 and Aldh1a3 - but the phylogenetic relationships and origins of these genes remained uncertain [59] , [65] . To identify gene gains and losses , one must first reconstruct the evolutionary genomic history of a gene family . We undertook a combination of phylogenetic and comparative genomic analyses of conserved syntenies that clarified the evolutionary history of the Aldh1a family . Phylogenetic results showed that Aldh1a1 and Aldh1a2 form sister clades and Aldh1a3 occupies a basal position in the phylogenetic tree rooted on cephalochordate Aldh1a genes ( Figure 1 ) . This analysis breaks the trichotomy observed in one previous analysis [59] and is opposite to the topology rooted on the far more distant Aldh2 gene family in another analysis [65] . Further support for the new understanding of Aldh1a family member relationships ( ( Aldh1a1 , Aldh1a2 ) Aldh1a3 ) comes from comparative genomic analyses of conserved syntenies in the genomic neighborhoods of Aldh1a paralogs in human , mouse , zebrafish , stickleback and medaka , which showed extensive conservation of syntenies between Aldh1a1 and Aldh1a2 genetic neighborhoods ( Figure 2B ) . The congruency of the inferred historical relationships that arise from the new phylogeny and conserved syntenies , which are independent datasets , forces the conclusion that Aldh1a1 and Aldh1a2 are sisters and both are cousins to the Aldh1a3 gene . Based on results obtained from the analysis of synteny conservation of the Aldh1a1 genomic neighborhoods across human and model organism genomes , we infer an evolutionary model that reconstructs the genomic history of the Aldh1a family , and integrates previous work by Nakatani et al . , ( 2007 ) [26] that had reconstructed the re-organization of the ancestral chromosomes ( named A to J ) of the last common ancestor of vertebrates through R1 , R2 and R3 genome duplications ( Figure 7 ) . Because Aldh1a2 and Aldh1a3 are syntenic ( on the same chromosome ) in human , zebrafish , and stickleback genomes , we conclude that this was the state in their last common ancestor ( Figure 7 step 1 ) . According to Nakatani's reconstruction , Hsa15 mostly derives from the post-R2 ancestral chromosome “A4” , which allows us to infer that Aldh1a2 and Aldh1a3 were syntenic in the ancestral chromosome A4 ( Figure 7 step 1 ) . After our comparative analysis of synteny conservation between human and mouse , which ruled out the possibility of reciprocal Aldh1a3 paralog losses ( Figure 3 ) and showed that Aldh1a3 genes are actual orthologs ( Figure 1 ) , we conclude that the Aldh1a3-ogm was already absent in the last common ancestor of tetrapods and teleosts ( Figure 7 step 1 ) . If Aldh1a2 and Aldh1a3 were syntenic in the ancestral state , we reason that a chromosomal translocation might have occurred during the evolution of the rodent lineage to separate them into different chromosomes ( e . g Mmu9 and Mmu7 in Figure 7 step 2 ) . Because the fourth Aldh1a paralog of rodents ( i . e . Aldh1a7 in mouse ) is adjacent and oppositely oriented to Aldh1a1 , separated only by 0 . 5 Mb with no intervening genes , we conclude that the fourth Aldh1a rodent paralog originated by a rodent-specific tandem gene duplication associated with a local inversion ( Figure 7 step 2 ) that was probably followed by subsequent amino acid sequence changes that destroyed its ability to synthesize RA [64] . In contrast to tetrapods , teleosts lack an Aldh1a1 ortholog , and whether this is due to a gene loss in teleosts , or a gene gain by tetrapods was previously unknown . Our whole-genome comparisons of conserved synteny answer this question by identifying genomic neighborhoods orthologous to the human ALDH1A1 genomic neighborhood in zebrafish , stickleback and medaka ( Figure 5 ) . This finding is consistent with the new Aldh1a phylogeny ( Figure 1 ) and provides strong evidence supporting the conclusion that Aldh1a1 was present in the last common ancestor before the tetrapod and teleost lineages split ( Figure 7 step 1 ) . Thus , we conclude that the absence of Aldh1a1 in teleosts is due to gene loss , probably in stem teleosts or ealier in stem actinopterygians ( Figure 7 step 3 ) , and discards the hypothesis that Aldh1a1 is a tetrapod innovation . This finding illustrates the power of comparative genomics to discern cases of gene losses from cases of gene gains , even in situations in which no proper outgroup is available . In human and mouse , Aldh1a1 and Aldh1a3-ogm genomic neighborhoods are not syntenic ( e . g . on Hsa9 and Hsa5 , respectively ) . Interestingly , however , in zebrafish , stickleback and medaka , genomic neighborhoods orthologous to those of human ALDH1A1 and ALDH1A3-ogm are syntenic ( e . g . , on Dre5 , GacXIII and Ola9 , see Figure 4A–C and right panels of Figure 7 ) . Thus , just as Aldh1a2 and Aldh1a3 were syntenic after R2 , it is likely that Aldh1a1 and Aldh1a3-ogm were also syntenic before the tetrapod-teleost split ( Figure 7 step 1 ) . This reasoning lead us to conclude that there might be a chromosomal translocation event that separated the Aldh1a1 and Aldh1a3-ogm genomic neighborhoods to two different chromosomes during the evolution of the tetrapod lineage ( Figure 7 step 4 ) . This predicted translocation event is supported by the reconstruction of ancestral chromosomes made by Nakatani et al . ( 2007 ) , [26] , in which a post-R2 ancestral chromosome named “A0” split into two main pieces that are today on Hsa5 and Hsa9 . Thus , we conclude that Aldh1a1 and Aldh1a3-ogm were syntenic in the ancestral chromosome A0 ( Figure 7 step 1 ) , which broke apart in tetrapods ( Figure 7 step 4 ) but remained intact in the teleost lineage ( Figure 7 step 3 ) . Consistent with our finding that Hsa1 and Hsa9 are related to the ALDH1A genomic neighborhoods in the human genome despite their lack of ALDH1A genes ( Figure 2A yellow boxes ) , Nakatani's reconstruction also shows that most of Hsa1 and Hsa19 derive from “A2–A5” and “A1–A3” , respectively , which are the other post-R2 homeologs derived from the ancestral chromosome “A” present in the genome of the last common pre-R1 vertebrate ancestor . Therefore , we conclude that the conserved synteny related to ALDH1A we detected in Hsa1 , Hsa5 , Hsa9 , Hsa15 and Hsa19 originated by R1 and R2 from the ancestral chromosome “A” in the genome of the last common ancestor of vertebrates . In Figure 7 , we propose two hypotheses to explain how a single gene located in pre-duplication chromosome “A” generated Aldh1a2 and Aldh1a3 in ancestral chromosome “A4” , and Aldh1a1 and Aldh1a3-ogm in ancestral chromosome “A0” inferred after R2 ( Figure 7 step 1 ) . The first hypothesis suggests Aldh1a duplication before R1 ( pink box in Figure 7 ) , and the second hypothesis invokes a translocation either before or after R2 ( tan box in Figure 7 ) ( see legend in Figure 7 for details ) . Independently of the order of events , however , both scenarios agree that the first duplication generated Aldh1a1/2 and Aldh3/3-ogm ancestral genes from the precursor Aldh1a1/2/3/3-ogm gene in the ancestral chromosome “A” . Because the available genomic databases of basally divergent vertebrates such as cartilaginous fishes ( e . g . dogfish , little skate or elephant shark ) , or from basally divergent craniates ( e . g . lampreys or hagfish ) , are still too fragmented to perform a comprehensive analysis of conserved synteny , testing the hypothetical “pre-R1 duplication” or “translocation” scenarios must be delayed . Supporting the postulated R3 teleost-specific genome duplication , our analysis of conserved synteny between the ALDH1A genomic neighborhoods and teleost genomes ( Figure 4 ) revealed that fish orthologs of human ALDH1A neighbors are mostly confined to two main chromosomes in each fish species , and no extra R3-generated aldh1a ohnologs have been preserved in duplicated copies ( Figure 7 step 5 ) . Analysis of conserved synteny ( Figure 4 ) supports the conclusion that each preserved duplicated Aldh1a gene is an actual ortholog of its partners within teleosts , and no evidence supports the complementary loss of aldh1a paralogs after R3 in different teleost lineages . The distribution of conserved co-orthologs in teleost paralogons , however , was asymmetric . In each of four genomic regions for three teleost species , one homeolog ( the primary chromosome ) conserved substantially more genes in the observed region than the other chromosome ( the secondary chromosome ) ( Figure 4D ) . This asymmetric distribution of syntenic gene conservation appears to be a common characteristic for R3-generated genomic neighborhoods , in agreement with previous observations of the analysis of the hox and parahox genomic neighborhoods in teleosts [17] , [89]–[91] and the analysis of syntenic blocs formed following tetraploidy in Arabidopsis [92] . Evolutionary sequence divergence among paralogs also often display asymmetry , with one paralog evolving at a rate similar to its tetrapod ortholog and the other paralog evolving at an accelerated rate , suggesting neofunctionalization [39] , [93]–[95] . During the analysis of the hox cluster it was noted that the fastest evolving hox genes belong to clusters that tend to lose their hox genes faster [89] , [96] . Furthermore , the asymmetric distribution of synteny conservation between parahox cluster paralogons in teleosts , was accompanied by asymmetric accumulation of introns and repetitive DNA elements in type III RTK genes , and asymmetric conservation of potential regulatory elements [91] . Thus , our observation of asymmetric chromosomal distribution of surviving co-orthologs in the aldh1a genomic neighborhoods extend previous observations in the hox and parahox genomic regions , to genomic neighborhoods with a great variety of gene types , suggesting that the probability of duplicate gene preservation depends not only on inherent evolutionary forces depending on gene function ( i . e . subfunctionalization and neofunctionalization ) , but also on properties pertaining to the architecture of the local genomic neighborhood . The R3 Aldh1a-ogm genes appear to represent cases in which , once gene organization had become altered in one of the duplicated regions , constraints that preserve genes became more relaxed , and therefore the chances of additional gene losses and further chromosomal rearrangements in the secondary chromosome were increased . At least two possible mechanisms could explain asymmetric co-ortholog retention: first , enhancers shared or embedded in genes at distant sites [70] , [91] , or second , epigenetic regulatory mechanisms based on chromatin architecture [71] , [72] . In principle , shared or distant enhancers or epigenetic regulatory signals must be retained in one homeolog , thus facilitating neighborhood gene retention , but can be lost from the other , allowing more gene loss and more rapid gene evolution due to greater relaxation of evolutionary constraints . In addition to the loss of aldh1a1 in stem teleosts ( Figure 7 step 3 ) , our genomic surveys revealed that aldh1a3 is absent from the genomic database of medaka fish ( Figure 7 step 6 ) . Identification of a genomic neighborhood in medaka that shows conserved orthologous synteny with the stickleback and human Aldh1a3 genomic neighborhoods ( Figure 6 ) provides evidence that aldh1a3 was lost in the medaka lineage after it diverged from the stickleback lineage ( Figure 7 step 6 ) . This finding illustrates again the power of comparative analysis of conserved synteny to provide evidence of gene loss . The finding of an apparent LTR-retrotransposon in the orthologous position occupied by aldh1a3 in stickleback and human suggests that the insertion of this mobile element may have caused the loss of aldh1a3 in medaka . Genomic data from medaka species phylogenetically close to Oryzias latipes is not yet available to more narrowly define the timing of this insertion event . The finding of the loss of aldh1a3 in medaka makes this organism the first known vertebrate with a single surviving Aldh1a paralog ( i . e . aldh1a2 ) , and made us wonder about the functional implications of gene loss . As a measure of gene function , consider expression patterns of Aldh1a genes . In the developing retina of mouse , frog , zebrafish and medaka , Aldh1a genes are expressed in a dorsal sector and in a ventral sector at the completion of optic cup invagination ( about E11 . 5 in mouse , stage 35 in frog , and 1 . 5 days post fertilization in zebrafish and medaka; Figure 8A ) . Different vertebrates express different Aldh1a genes in different dorso-ventral sectors of the eye . The right column of Figure 8 summarizes the main expression patterns of the Aldh1a family in the retina of different animals ( Aldh1a1 in red , Aldh1a2 in blue , and Aldh1a3 in green ) . Aldh1a paralogs expressed in the dorsal sector of the retina include Aldh1a1 ( but not Aldh1a2 ) in mouse; both Aldh1a1 and Aldh1a2 in frogs and birds ( e . g . chicken and quail , not included in Figure 8A for simplicity ) ; and Aldh1a2 ( but not Aldh1a1 ) in teleosts ( e . g . zebrafish and medaka ) . The main Aldh1a paralog expressed in the ventral sector of the retina is Aldh1a3 both in tetrapods ( e . g . mouse , frog and birds ) and in at least one teleost ( e . g . zebrafish ) . In contrast , in medaka , which lacks an aldh1a3 paralog , we found strong expression of aldh1a2 ventrally ( Figure 6 ) . Dotted regions depict weak expression of Aldh1a genes in a small part of each dorso-ventral sector or from earlier developmental stages prior to the complete invagination of the optic cup in Figure 8A . The rules of ancestral reconstruction imply that the retina of the last common vertebrate ancestor probably had a dorsal and a ventral sector , and the original Aldh1a1/2/3/3-ogm gene prior to the expansion of the Aldh1a family gene was likely expressed in both dorsal and ventral sectors ( Figure 8A step 1 ) . According to the evolutionary model proposed in Figure 7 , the first expansion of the Aldh1a family occurred before R2 and generated Aldh1a1/2 and Aldh1a3/3-ogm . Because Aldh1a3 is the major paralog in the ventral sector of the retina in extant tetrapods and teleosts , and because Aldh1a1 or Aldh1a2 are the major paralogs in the dorsal sector of the retina in both tetrapods and teleosts , we infer that after the first duplication prior to R2 , Aldh1a1/2 inherited the subfunction leading to expression in the dorsal sector of the retina , and Aldh1a3/3-ogm inherited the subfunction causing expression in the ventral sector ( Figure 8A step 2 ) . It is probable that this subfunctionalization event contributed to the preservation of both paralogs as expected under the duplication-degeneration-complementation ( DDC ) model , in which the summation of the subfunctions that were split between gene duplicates equals the ancestral function prior the duplication event [3] . After R2 , Aldh1a3-ogm was lost and Aldh1a3 became the main ventral source of RA in the retina . On the other hand , both Aldh1a1 and Aldh1a2 retained expression in the dorsal sector because it is preserved in frog , chicken and quail , but not in mouse . Thus we conclude that the absence of Aldh1a2 dorsal expression in mouse retina is due to a loss of an ancestral expression domain , which can be interpreted as an evolutionary innovation due to late subfunction partitioning [3] , in which a function that was originally possessed by both Aldh1a1 and Aldh1a2 became partitioned exclusively to Aldh1a1 ( Figure 8A step 3 ) . Analysis of the ALDH1A2 expression pattern in the human retina will help elucidate whether the loss of the Aldh1a2 dorsal expression domain occurred before the split of lineages leading to human and rodents , or if it is a feature that has been acquired specifically in the rodent lineage . An important question is how gene loss can impact the evolution of gene regulation and gene function in surviving paralogs . After the loss of Aldh1a1 in teleosts , Aldh1a2 became the only source of RA in the dorsal retina , taking full responsibility for subfunctions originally shared with Aldh1a1 . Natural selection would have gradually increased the strength of the ancestral dorsal domain of Aldh1a2 ( Figure 8A step 4 ) . Medaka lacks both aldh1a1 and aldh1a3 orthologs , and the only surviving Aldh1a gene is aldh1a2 , which is expressed in both the dorsal and ventral domains of the retina ( Figure 8A step 5 ) . The fact that in zebrafish and mouse , Aldh1a2 is expressed early in the ventral retina prior to the closure of the optic cup and becomes progressively down-regulated until the completion of optic cup invagination ( arrow in Figure 6B ) [88] , suggests that early expression followed by down-regulation of Aldh1a2 is an ancestral feature and that medaka evolved an innovative heterochronic mechanism to avoid the ventral down-regulation of aldh1a2 and to increase its ventral expression at later stages . Thus , it is likely that the dorsal and ventral paracrine sources of RA that have been suggested to regulate the development of perioptic mesenchimal derivative structures [56] is an ancestral feature that might be still preserved in teleosts . Comparative and functional analysis of the regulation of aldh1a paralogs during the development of the eye and other tissues in medaka , zebrafish and in other fishes , particularly outgroups , will be required to test this hypothesis . The evolution of functions among Aldh1a paralogs illustrates what may be a general phenomenon associated with evolution after genome duplication: gene loss without altering developmental programs due to the preservation of functions in surviving paralogs . In our case study , the unaltered ancestral program provides both a dorsal and ventral supply of Aldh1a enzyme and hence dorsal and ventral sources of RA during retinal development . Comparative analysis shows that different paralogs can perform equivalent functions in different species . For instance , the ventral sector of the retina expresses aldh1a2 in medaka and aldh1a3 in zebrafish; and the dorsal sector of the retina expresses Aldh1a1 in mouse and aldh1a2 in zebrafish . Similar cases of what has been called function shuffling have been described for Hox genes [42]; Bmp genes [97] , and Twist genes [43] . Gitelman ( 2007 ) proposed the term synfunctionalization to describe the process by which a paralog acquires the expression pattern of another paralog by gaining new regulatory elements , and thereby allowing losses of genes without changing the ancestral developmental program [43] . The acquisition of enhanced ventral expression by aldh1a2 in the face of aldh1a3 loss in medaka suggests several possible mechanisms for the apparent shuffling of functions between aldh1a3 and aldh1a2 that do not require the evolutionary gain of new regulatory elements ( Figure 8B ) . Based on our findings , we propose a general mechanistic model to explain the loss of a paralog without altering the ancestral developmental program . After gene duplication from an ancestral gene a/b ( Figure 8B Step 1 ) , paralog b ( e . g . aldh1a3 ) could lose the dorsal subfunction without penalty ( Step 2 ) because it is covered by paralog a ( e . g . aldh1a2 ) . Next , mutations in negative regulatory elements or in upstream negative regulators that normally down-regulate paralog a expression in later developmental stages ( e . g . , after retina cup invagination ) would facilitate an innovative heterochronic paralog a expression ( Step 3 ) . Finally , natural selection or genetic drift could act on natural variation that positively strengthens paralog a expression in the ventral domain ( Step 3 ) , while allowing relaxed selection for paralog b expression ( Step 4 ) , thereby facilitating the loss of paralog b ( Step 5 ) without loss of the ancestral developmental program ( Step 6 ) . Overall , our results illustrate how comparative genomic analyses of conserved synteny , coupled with reconstruction of ancestral chromosomes , can provide a phylogenetic framework necessary for the identification of lineage-specific gene losses . Our analysis provides evidence for early subfunctionalization and late subfunction-partitioning , and for the acquisition or modification of subfunctions by surviving paralogs that preserve unaltered ancestral developmental programs in the face of gene loss . Understanding the evolution of gene functions is fundamental for the proper interpretation of comparative analyses , especially when using model organisms to understand human gene functions . In the case of the Aldh1a family , although RA is important in human disease , we still know little about the spatio-temporal dynamics of the expression domains and functions of ALDH1A1 , ALDH1A2 and ALDH1A3 genes during human development and adult organ homeostasis , other than RT-PCR studies [98] , which do not provide enough resolution at the single cell level to understand how the sources of RA regulate physiological action . The evolutionary framework defined here provides information essential for the functional connectivity of human and model organism genomes , not only for RA signaling in eye development , but for the many organs in which RA plays important functions , including axis and limb development and cancer biology .
All animals were handled in strict accordance with good animal practice as defined by the relevant animal welfare bodies , and all animal work was approved by the University of Oregon Institutional Animal Care and Use Committee ( A-3009-01 , IACUC protocol #08-13 ) . Alignments of ALDH1A proteins from vertebrates and cephalochordates were generated with clustalX [99] and corrected by eye . Only the unambiguous part of the alignment was considered for phylogenetic tree reconstructions ( Figure S1 provides sequence alignments ) . The ProtTest tool was used to choose the best-fit protein evolutionary model [100] , resulting in the LG+I+G [101] and the JTT+I+G [102] as the top two selected , with a relatively low value of deltaAIC = 92 . 92 ( AIC = 18797 . 45 and 18890 . 37 , respectively ) . Because different phylogenetic methods have different limitations [103] , we compared results from four phylogenetic approaches: i ) Bayesian phylogenetic inferences were calculated with MrBayes [104] , using the JTT model as well as a gamma distribution for rate variation ( divided into four categories ) and a proportion of invariant sites . We ran two chains for 5 million generations , sampling every 100 iterations with a 25% burn-in . ii ) Maximum-likelihood ( ML ) analysis was conducted using PHYML [105] , with an LG+I+G and JTT+I+G model . The alpha parameter of the gamma distribution ( 1 . 41 ) and the proportion of invariable sites ( 0 . 19 ) were estimated from the sample , considering four categories of substitution rates . The topology , branch lengths , and rate parameters of the tree were required to be optimized . iii ) Maximum-parsimony ( MP ) analysis ( MEGA package , [106] used the close-neighbor-interchange approach with one level of search , and added 10 replicas of random trees , and 100 replications to calculate the bootstrap value that supports each node of the tree . iv ) Neighbor-joining phylogenetic ( NJ ) tree ( MEGA package , [106] was inferred taking into account among-site rate heterogeneity with four gamma-distributed categories . This approach has been previously shown to provide equivalent results to those obtained by ML under conditions of low sequence divergence , with the advantage of a low computing-time cost [107] . The alpha parameter 1 . 41 was estimated from the sample using PHYML under a JTT substitution model . Concordance of trees from each of the different methods , bootstrap proportions and posterior probability estimates were used to examine the robustness of nodes . Aldh1a1/2/3 proteins predicted from gene sequence data from the cephalochordate Branchiostoma floridae were used to root the phylogenetic tree of the vertebrate Aldh1a family . Tunicate Aldh1a1/2/3 proteins were not included to avoid possible artifacts arising from long branches shown previously for Aldh genes [59] . The automatic tools developed by Catchen et al . [48] to detect synteny conservation allowed us to perform comprehensive genomic comparisons between the human genome and other fully or partially assembled genomes from a wide variety of model organisms . These automatic tools use a reciprocal best hit BLAST analysis pipeline to define groups of paralogy between a primary genome and an outgroup genome . For instance , when the human genome is compared with outgroup genomes that diverged prior the two rounds of genome duplication R1 and R2 ( i . e . the urochordate Ciona intestinallis or the cephalochordate Branchiostoma floridae in Figure 2A ) , each human gene will belong to a group of paralogy that is anchored to a gene from the outgroup genome . Use of multiple outgroup genomes and merging clusters anchored by outgroup paralogs help to minimize errors derived from the automatic reciprocal best hit BLAST pipeline due to the effect of losses , duplications or sequence divergence of outgroup genes ( for details on best hit BLAST pipeline analysis , see [48] ) . Dotplots graphically represent the distribution of paralogous genes ( crosses ) within the primary genome ( e . g . Figure 2A ) , or the distribution of orthologous genes between the primary and outgroup genomes ( e . g . Figure 3A ) , using the results generated with the automatic BLAST analysis pipeline . In the case of an orthology dotplot , genes belonging to a selected chromosome in the outgroup are displayed along the x-axis of the plot in the order they appear in that genome . Orthologs of those genes are displayed on their respective chromosomes in the primary genome directly above the location of the gene on the selected chromosome in the outgroup , not in their order in the second genome . Scaled dotplots represent a variant in which the paralogs ( or orthologs ) of genes on the selected chromosome are displayed according to their natural chromosomal positions in the genome ( e . g . Figure 2A ) . For instance , given an orthologous dotplot with Danio rerio as the primary genome and human as the outgroup ( Figure 4A ) , each two paralog genes originated by R3 in Danio will be aligned above their human ortholog on the x-axis . Composite dotplots overlap multiple dotplots from the analyses of various regions of interest ( crosses labeled with different colors ) and different outgroup genomes ( e . g . Figure 2A ) . Circleplots represent user-selected chromosomes as arcs along the circumference of a circle . The origins of lines connecting positions along the arcs represent pairs of paralogous genes within the same species ( Figure 2D–F ) or orthologous genes between two different species ( Figure 3B ) . Gene loci that are close to each other may appear overlapped as single crosses in the dotplot or a single connecting line in circle-plots due to the selected graph resolution . Clusters in the Synteny Database were created by coupling results from the reciprocal best hit BLAST pipeline with the use of a sliding window analysis that links chromosome segments with conserved synteny ( for details see [48] ) . Clusters that link chromosomal segments within the same species represent paralogous syntenic conservation ( e . g . Figure 2B–C ) , and clusters that link chromosomal segments between different species represent orthologous syntenic conservation ( e . g . Figure 4E–G ) . The Synteny Database provides clusters produced using several different sliding window sizes measured in terms of contiguous gene number . Smaller window sizes identify tightly-conserved syntenic regions in which gene order and orientation are well preserved while larger window sizes can accommodate chromosomal rearrangements ( inversions , transpositions , translocations , and small duplications ) . The Synteny Database is especially useful to provide evidence of ohnologs gone missing ( ogm ) by uncovering the putative chromosomal region that still preserves paralogous syntenic conservation , but lacks a certain ohnolog of interest ( e . g . Figures 5 and 6 ) . Full coding sequence of aldh1a2 cDNA from Medaka Oryzias latipes ( Cab strain ) and the aldh1a3 cDNA from zebrafish Danio rerio were cloned after being amplified from cDNA by PCR with specific primers designed from genomic scaffold sequence data ( medaka: 200506-scaffold21 and zebrafish: Zv5Scaffold1492 and NA2068 ) ( Ola1a2F: 5′ATGACTTCCAGTAAGATCGAGATCCC3′ and Ola1a2R: 5′CATTAACGTTTCATCCATTACTGTCC3′; Dre1a3F: 5′GTCCACACAATAATCTACTCTACAGC3′; Dre1a3R 5′CATATGTTTGCGCTTAGCTGCCATG3′ ) . Full length cDNA sequences were submitted to GenBank ( medaka aldh1a2: FJ516380 , and zebrafish aldh1a3: DQ300198 ) . A zebrafish adh1a2 clone [86] , a clone containing a zebrafish aldh1a3 800 nt-fragment from exon 7 to exon 13 ( cloning primers: 5′GGAGCTGCGATCGCTGGTCACATG3′ and 5′CTGAGTTTGATAGTGATGGCTTTGAC3′ ) , and a clone containing a medaka aldh1a2 527-nt fragment from exon 12 ( cloning primers: 5′GGAGGATACAAAATGTCTGGGAATGG3′ ) to the 3′UTR ( 5′CATTAACGTTTCATCCATTACTGTCC3′ ) were used to synthesize riboprobes for whole-mount in situ hybridization using standard procedures [108] , [109] , with slight variations: NBT and BCIP were used instead of the BM purple . | Gene duplication may facilitate the acquisition of genetic diversity . Little is known , however , about the impact of gene loss on the functions of surviving genes . When a gene is lost , can other closely related genes evolve to perform the functions of the lost gene ? Answering this question can be difficult because the proof for gene loss is based on negative evidence and thus can easily pass unnoticed . Here , we illustrate how the comparison of genomic neighborhoods in different species can help reconstruct the chromosomal history of a gene family and provide robust evidence for gene loss , even without an appropriate early-diverging comparator group . Identifying gene loss is important because it helps distinguish between gene gain as a lineage-specific innovation and gene loss as a lineage-specific simplification . As a case study , we investigated the expression of the Aldh1a family , which is crucial for retinoic acid signaling in development of eyes , limbs , the brain , and in cancer . Results showed that gene loss is indeed associated with the evolution of functional change in surviving gene family members . Our results highlight the relevance of comparative genomics for identifying gene loss and improving the functional connectivity among human and model organism genomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/comparative",
"genomics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"developmental",
"biology/developmental",
"evolution",
"developmental",
"biology/pattern",
"formation",
"genetics",
"and",
"genomics/gene",
"function",
"computational",
"biology/genomics",
"developmental",
"biology/molecular",
"development"
] | 2009 | Consequences of Lineage-Specific Gene Loss on Functional Evolution of Surviving Paralogs: ALDH1A and Retinoic Acid Signaling in Vertebrate Genomes |
Only a few studies have investigated the potential of using geotagged social media data for predicting the patterns of spatio-temporal spread of vector-borne diseases . We herein demonstrated the role of human mobility in the intra-urban spread of dengue by weighting local incidence data with geo-tagged Twitter data as a proxy for human mobility across 45 neighborhoods in Yogyakarta city , Indonesia . To estimate the dengue virus importation pressure in each study neighborhood monthly , we developed an algorithm to estimate a dynamic mobility-weighted incidence index ( MI ) , which quantifies the level of exposure to virus importation in any given neighborhood . Using a Bayesian spatio-temporal regression model , we estimated the coefficients and predictiveness of the MI index for lags up to 6 months . Specifically , we used a Poisson regression model with an unstructured spatial covariance matrix . We compared the predictability of the MI index to that of the dengue incidence rate over the preceding months in the same neighborhood ( autocorrelation ) and that of the mobility information alone . We based our estimates on a volume of 1·302·405 geotagged tweets ( from 118·114 unique users ) and monthly dengue incidence data for the 45 study neighborhoods in Yogyakarta city over the period from August 2016 to June 2018 . The MI index , as a standalone variable , had the highest explanatory power for predicting dengue transmission risk in the study neighborhoods , with the greatest predictive ability at a 3-months lead time . The MI index was a better predictor of the dengue risk in a neighborhood than the recent transmission patterns in the same neighborhood , or just the mobility patterns between neighborhoods . Our results suggest that human mobility is an important driver of the spread of dengue within cities when combined with information on local circulation of the dengue virus . The geotagged Twitter data can provide important information on human mobility patterns to improve our understanding of the direction and the risk of spread of diseases , such as dengue . The proposed MI index together with traditional data sources can provide useful information for the development of more accurate and efficient early warning and response systems .
Dengue has become a major concern for public health authorities in tropical and sub-tropical developing countries [1]; the frequency and magnitude of epidemics , the incidence of severe disease , and the rate of hospitalizations have increased in the past few decades [2] . Asia-Pacific countries bear the heaviest disease burden of dengue where over 1·8 billion people are estimated to be at risk of infection [3 , 4] . Dengue also poses a serious economic challenge to countries due to high costs of dengue prevention and control programs , particularly during epidemic peaks [5 , 6] . Timely and accurate disease reporting and forecasting is the pillar of infectious disease control . However , public health agencies often report disease trends and outbreaks with severe delays , and reporting tends to be based on aggregated disease data at national or regional levels with little information about disease counts and trends at local levels . Dengue is a notifiable disease in most endemic countries; however , several studies revealed high levels of under-reporting in routine surveillance systems , particularly from ambulatory care settings [2] . These shortcomings hamper programmatic efforts on the ground to mount timely , context-specific , and effective response to abnormal disease events , including incipient epidemics [7] . Population growth , unplanned urbanization , increased vector density , and climate variability are all identified as important contributing factors to dengue propagation [8] . Spatial and temporal variation in interactions among hosts , dengue viruses , vectors and the environment have led to a heterogeneous distribution of dengue risk across geographical locations [9–11] . Understanding how these complex interactions influence the epidemiology of dengue at different spatial and temporal scales is important to assess transmission risk and allocate resources efficiently [8 , 12] . A main obstacle to studying such complex interactions has been the limited availability of large-scale spatial and temporal datasets . Several studies have explored using near real-time streaming data from Twitter to investigate public health trends . As of the first quarter of 2017 , there were about 328 million monthly Twitter users worldwide [13] . This large volume of social media data may be exploited for public health monitoring and surveillance purposes [14 , 15] . The most recent literature has focused on analysing Twitter content using text mining methods to estimate and forecast infectious disease activity [16 , 17] , predict heart disease mortality [18] , and measure health-related quality of life [19] . One study explored the use of Twitter content for dengue forecasting , but focused on verifying the correlation between number of dengue cases and dengue-related tweets posted over the same time period [20] . In this study , we investigated the use of publicly available geotagged Twitter data for predicting the spatio-temporal clustering patterns of dengue incidence . First , we designed , implemented and evaluated an algorithm that harvested and analysed real-time Twitter streams to estimate proxies of human mobility in a densely populated urban area . Then we weighted the incidence of dengue in all neighborhoods by the mobility proxies to specific locations and generated a dynamic Mobility-weighted Incidence ( MI ) index . Lastly , we demonstrated that the MI index was highly predictive of the temporal and spatial patterns of dengue spread in Yogyakarta municipality , Indonesia .
The study was conducted in Yogyakarta municipality , one of the five districts and the capital of Yogyakarta Province in Indonesia . Yogyakarta municipality is a medium sized , densely populated , and rapidly developing urban area , spread over 32 . 5 km2 with an average population density of 14 , 000 persons/km2 . It is located about 538 km away from the capital Jakarta and lies between 75 to 132 m above sea level in the central southern part of Java island at 07°45’57”–07°50’25” S and 110°20’41”–110°24’14” E [21] . Yogyakarta municipality is divided into 45 neighborhoods ( Fig 1 , number 1 to 45 ) , ranging in surface area between 0 . 3 and 1 . 68 km2 . This study used neighborhoods as the geographical unit of observation . We obtained monthly dengue cases ( i . e . dengue fever , dengue haemorrhagic fever , and dengue shock syndrome ) for each neighborhood ( Den ) during the period August 2016–June 2018 from the Dengue Surveillance Report of the Yogyakarta Municipality Health Office . We complemented dengue surveillance data with geotagged tweets posted in the administrative boundaries of the study area during the same period . To achieve this , we employed the Twitter’s Application Programming Interface ( API ) and selected Tweets within Yogyakarta municipality for analysis . We only extracted the user identification string , timestamp , and longitude and latitude of the user’s location in the Tweets . We overlaid the geotagged tweets on the administrative map of the study area and exchanged the geocode to the neighborhood identification number ( ID ) . We formulated an algorithm to estimate a dynamic MI index , quantifying the level of exposure to virus in any given study neighborhood due to importation form other neighborhoods . The MI index was calculated based on Twitter users’ mobility patterns between pairs of neighborhoods . The mobility patterns were computed by estimating the rate with which a Twitter user in one study neighborhood re-tweeted in another neighborhood within the same month . Based on this information , we generated a monthly matrix ( It ) measuring the cumulative number of mobility events between each pair of neighborhoods at time t , in months . Then , we created the monthly mobility network ( Ni , j , t ) of neighborhoods by multiplying ( It ) with its transpose ( ItT ) . We set the diagonal of the 45×45 matrix of the affiliation to zero . Then we standardized the monthly mobility matrix ( Ni , j , t ) by dividing it by the total number of mobility events observed at time t ( Nt ) for all the neighborhoods . This ensured that , at a fixed time , the mobility matrix would always sum to 1 . We referred to the standardized mobility matrix as , Ňt . To capture the total exposure to incoming mobility into each neighborhood , j , we aggregated the standardized mobility from all the 45 neighborhoods over one month and referred to this as the TWjt . The TW index is thus a time dependent vector of length 45 . We further constructed a new matrix by multiplying the standardized mobility , Ňijt , by the vector of the number of dengue cases reported in each neighborhood i ( of outgoing mobility ) , and we referred to this index as importations and computed it as Iijt = Ňijt × Denit . Lastly , to capture the total exposure to the dengue virus imported due to human mobility into each neighborhood , j , from all the other neighborhoods , i , we aggregated the importations , Iijt , from all the 45 neighborhoods over one month and referred to this as the MIjt . The MI index is thus a time dependent vector of length 45 . We then investigated the association between dengue incidence and the Den , TW and MI variables using a Bayesian spatio-temporal modelling framework assuming a Poisson distribution of the monthly counts in each neighborhood . In the model , we estimated and adjusted for the spatial covariance between neighborhoods using an unstructured spatial covariance matrix . We further adjusted for the influence of population size variability across neighborhoods by offsetting population size . Thus , the regression analysis assessed predictors of the incidence of dengue . We implemented the models using the INLA R-package [22 , 23] . In the regression model , we started out by investigating how much of the variability in the dengue counts could be explained by the spatial covariance and intercept model only ( the null model ) , leaving out all predictor variables . Subsequently , we included the MI , TW and Den variables one lag at a time ( crude ) , and then all lags 1 to 6 months simultaneously , but only one variable at a time . For variables showing important prediction skill , we also analyzed their combined predictive ability . The models were fitted with all lags in the same model , but with only one of the MI , TW and Den variables at a time . The model structure can be described as: yit = Poisson ( λit ) ; λit = Eit ρit log ( ρit ) = ηit ηit = bo+ Σ βk zi ( t-k ) + ui+vi + log ( pi ) The terms ui and vi are the spatial effects , representing unspecified features of neighborhood i that do and do not display spatial structure [24] , respectively . The k indicates the lag in months and takes values from 1 to 6 . The z corresponds to the variables MI , TW and Den . The coefficient βk represents the regression coefficients for the variable z at lag k . The pi variable offsets the population size of neighborhood i . The models were evaluated based on the Bayesian Information Criterion ( BIC ) and the estimate of R-square , as well as on prediction performance according to the standardized root mean square error ( SRMSE ) .
The total number of dengue cases during the 23-month study period was 1 , 203 , with the highest monthly count of 13 cases reported for neighborhood ID 11 in August 2016 . The monthly incidence of dengue in the study area increased gradually from December to March of next year and then decreased until the start of the rainy season in October ( Fig 2 ) . Overall , the incidence of dengue was decreasing over the study period . The number of Twitter users and the population size of each study neighborhood are shown in S1 Fig . The monthly mobility patterns for the 45 neighborhoods appeared to be relatively consistent over the study period , except that a slight increase in the number of mobility events was observed over the same period . The mobility patterns varied considerably across different pairs of neighborhoods ( Fig 3 ) . The MI index ( Fig 4 ) for each neighborhood reflected a combination of the mobility estimates and the disease counts ( Fig 2 ) . In general , we found that the MI index was not only higher for the neighborhoods with relatively higher mobility to other neighborhoods , but also reflected the decreasing trend in the disease counts over time ( Figs 3 and 4 ) . Table 1 describes the crude and adjusted model estimates of the lag effects of Den , TW and MI using the Bayesian spatio-temporal regression model . We found that the mobility and the centrality of a neighborhood proved not to be important for predicting the incidence of dengue on its own . This is shown by comparing the model fit of the TW lag variables ( crude and adjusted ) to the null model and their observed lack of difference in the R-square , BIC and SRMSE in Table 1 . In contrast , the Den and MI variables provided important information for predicting the incidence of dengue at lead times 1 to 6 months based on the crude and adjusted estimates of the model ( Table 1 ) . The coefficients from the crude and adjusted models are graphically presented in Fig 5 . Unsurprisingly , the uncertainty and confidence intervals for the coefficient estimates increased in the lag adjusted models compared to the crude single lag models . Overall , the coefficients were smaller in the adjusted models . This is because of the similarity of information carried over in lags of a specific variable , i . e . due to temporal covariance . In the adjusted models , we observed a decreasing pattern in the association to the Den and MI variables with increasing lags , with the exception that both peaked at lag 3 months . While most lags associated with the Den variable showed statistically significant associations , the associations with the MI variable were more uncertain , with the exception of at lag 3 months . However , since the SRMSE was lower for the MI model , it appeared that this variable still included more vital information for predicting the incidence of dengue in the neighborhoods . Furthermore , an inspection of the crude estimates strongly supported this conclusion , where the MI variable at lag 3 months had clearly the best predictive ability and almost the same predictive ability as the adjusted models with all lags , in view of the R-square , BIC and SRMSE values ( Table 1 ) . The Den variable at lag 3 months did not show a similar good performance with significantly lower predictive ability , R-square , BIC and SRMSE . The model including both the Den and MI variables at lags 1 to 6 months estimated an R-square , BIC and SRMSE of 0 . 271 , 1140 . 8 and 0 . 778 , respectively , and showed considerably higher predictive ability compared to the adjusted models of the Den and MI variables alone ( Table 1 ) . This supports the fact that these variables contributed different information to the predictive ability of the model . Looking at the coefficients in this combined model , the estimates were not very different than those obtained from the adjusted single variable model estimates , confirming the exclusive unique contribution of these two variables to the predictive ability .
This study revealed insights into how the intra-urban outbreak risk relates to a combination of human mobility and the size of local outbreaks , and developed a new early warning variable indicating the risk of spread . The indicator integrated human mobility proxies derived through an analysis of Twitter user geolocation data with disease surveillance data , and demonstrated its ability as a predictor of dengue incidence up to 6 months lead time at the intra-city level . The proposed MI index captures dynamic network properties in a simplified and condensed form and can be used in regression models , similar to the model fitted here , to describe complex spatio-temporal interactions between human mobility and disease spread . We found that the impact of human mobility on disease spread cannot be effectively studied without combining mobility information with disease incidence data . This is not surprising because mobility does not necessarily translate into a greater exposure to the circulating virus unless it is combined with disease incidence information—this is exactly what the new MI index captures . We propose further the development of methods and the testing of the MI index , particularly for predicting the risk of incidence and spread of dengue with a lead time of 3 months . We also propose that future research should consider the combined effects of the MI index and the past cases in the same location ( the Den variable ) , which was found to contribute significantly to the prediction accuracy of the models . These findings have implications for empirical studies assessing the incidence risk ( such as adjusting for mobility bias in cluster randomized trials ) and for risk assessments at both micro and macro geographical levels , especially in the development of early warnings systems using near-real time data [25 , 26 , 27] . The demonstrated predictive ability of the MI index alone ( 20% of the variability in the incidence of dengue in mutually exclusive locations ) and in combination with auto-correlative terms ( 27% of the variability in the incidence of dengue in mutually exclusive locations ) hold great promise for improving predictions , early warning systems , and timely response . It also highlights the importance of understanding better the role of population mobility in the spread of arboviruses at the intra-city scale . The combined use of autoregressive terms and the MI index along with other factors , such as weather variability , environmental characteristics , and vector activity , is likely to yield substantially improved predictions . Furthermore , adjusting for virus exposure using the MI index would be important for studies mapping the spatial and spatiotemporal risk factors for dengue . For instance , human mobility , as shown in this study , is an important predictor and a potential confounder of the local incidence of dengue at the spatiotemporal scales . This analysis benefited from a novel data source and a novel procedure for tracking and predicting human mobility from publicly available social media data , providing a low-cost source of information . Given the high explanatory power of the MI index to describe the variability in dengue incidence , we believe that social media driven mobility indicators have the potential to allow researchers to assess the risk of communicable diseases , such as dengue , in real time by capturing dynamic network properties of importance for timely disease control . We estimated the user mobility patterns and the affiliation network in a relatively small but densely populated urban area by utilizing data from the Twitter’s API . The retrieved data from the API represent only about 1% of the Twitter volume , but previous research suggests that when geographic boundary boxes are used almost the complete sample of Twitter location data can be extracted [28 , 29] . Ideally , it is better to use data from Twitter’s Firehose . The major drawbacks of Firehose data are its prohibitively high cost and large storage and computational resource requirements [29] , both of which can adversely affect the sustainability of translational applications of such data for public health preparedness and response . We derived mobility from a rather short ( 23-month ) time-series data from August 2016 to June 2018 to infer for the degree of association between the MI index and the observed dengue cases . Future studies should assess the predictive performance of the MI index further by using longer prospective validation series and building more complete models of dengue disease dynamics by including other predictive factors . We further suggest that future studies investigate non-linearities in virus exposure and response relationships and implement a distributed lag approach . Despite these limitations , we were able to demonstrate a strong association of the MI index with reported dengue cases . Therefore , we believe that the MI index holds promise as an alarm variable in disease surveillance and early warning systems , contributing to a better understanding of spatial patterns of outbreak clusters over time , namely dynamic hotspots . A limitation of this study is the assumption that user movements between consecutive tweets were representative of the overall population mobility , while in fact Twitter users may represent a selected group of individuals . It is , however , important to note that the use of Twitter and other social media platforms is very common in Indonesia [30] , and that the demonstrated predictive ability of the MI index in this study supports the belief that Twitter data can capture the important aspects of mobility relevant for the spread of dengue in a densely populated urban area . This goes hand in hand with prior studies validating Twitter as a viable data source to study human mobility [31 , 32] . Using mobile phone data with geo-tags would have been a better alternative , although the downside is that such data are harder to acquire and use prospectively over time . Yet , human mobility patterns extracted from geotagged tweets have been reported to have similar overall features with mobile phone records [33] . The analysis employed a novel procedure for tracking and predicting human mobility and dengue spread at the intra-urban level using publicly available social media data from Twitter . We demonstrated that dengue cases were well predicted by a dynamic mobility-weighted incidence index at a lead time of 1 to 6 months at the within-city level . The newly developed MI index captures the micro-level dynamics of human mobility and virus importation in a condensed form , making it useful for use in empirical regression models . The results suggest that human mobility is an important driver of the movements of incidence clusters within a city . We conclude that this novel early warning indicator has implications for dengue surveillance and early warning systems and can potentially enhance timely decision-making and coordination within the public health system . | Recent studies have shown that Twitter can be utilized as a tool for health research , and aggregated large-scale social media data can indicate the risk of infectious disease in real-time with high accuracy and at low cost . However , most of these studies relied primarily on content analysis or text mining , while only a few analyzed the networks of Twitter users . None has incorporated user geolocation data to explain health outcomes at an intra-urban level . Currently dengue early warning systems rely on syndromic surveillance , which lacks completeness and timeliness . Effective syndromic surveillance is rarely achieved due to its technical complexity and a general lack of capacity . Researchers have assessed vector indices , meteorological factors and environmental variables as predictors of dengue incidence , but have failed to capture the complexity of transmission as it relates to human behaviors and movements . Here we develop an algorithm to estimate a dynamic mobility-weighted incidence index ( MI ) , which quantifies the level of exposure to virus importation in a given neighborhood . The proposed index is based on publicly available social media and routine disease surveillance data , and provides a low-cost source of information for assessing the risk of spread of communicable diseases , such as dengue . This study suggests that the MI index is of utility and significance for dengue surveillance and early warnings systems and can enhance timely decision-making within the public health system . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2019 | A combination of incidence data and mobility proxies from social media predicts the intra-urban spread of dengue in Yogyakarta, Indonesia |
The use of internet search data has been demonstrated to be effective at predicting influenza incidence . This approach may be more successful for dengue which has large variation in annual incidence and a more distinctive clinical presentation and mode of transmission . We gathered freely-available dengue incidence data from Singapore ( weekly incidence , 2004–2011 ) and Bangkok ( monthly incidence , 2004–2011 ) . Internet search data for the same period were downloaded from Google Insights for Search . Search terms were chosen to reflect three categories of dengue-related search: nomenclature , signs/symptoms , and treatment . We compared three models to predict incidence: a step-down linear regression , generalized boosted regression , and negative binomial regression . Logistic regression and Support Vector Machine ( SVM ) models were used to predict a binary outcome defined by whether dengue incidence exceeded a chosen threshold . Incidence prediction models were assessed using and Pearson correlation between predicted and observed dengue incidence . Logistic and SVM model performance were assessed by the area under the receiver operating characteristic curve . Models were validated using multiple cross-validation techniques . The linear model selected by AIC step-down was found to be superior to other models considered . In Bangkok , the model has an , and a correlation of 0 . 869 between fitted and observed . In Singapore , the model has an , and a correlation of 0 . 931 . In both Singapore and Bangkok , SVM models outperformed logistic regression in predicting periods of high incidence . The AUC for the SVM models using the 75th percentile cutoff is 0 . 906 in Singapore and 0 . 960 in Bangkok . Internet search terms predict incidence and periods of large incidence of dengue with high accuracy and may prove useful in areas with underdeveloped surveillance systems . The methods presented here use freely available data and analysis tools and can be readily adapted to other settings .
Google has reported success in using terms entered into its search engine ( www . google . com ) to predict trends in Influenza-Like Illness ( ILI ) cases one to two weeks ahead of the US CDC Morbidity and Mortality Weekly Report [1] . Several studies have reported similar results for influenza surveillance using Google search data , Yahoo search data , and internet advertising [2]–[8] . The Google research indicates that as the weekly incidence of influenza increases or decreases , the volume of certain internet search terms within the same geographical region change with a high level of correlation and predictability . Using a near real-time ability to collect search data ( within 24 hours as opposed to one to two week lead time for US CDC reporting ) , the researchers were able to obtain information on the trend of ILI patterns in a more timely fashion than traditional surveillance . Though the first efforts to use search terms from www . google . com have focused on influenza , this pathogen may be one of the more difficult to predict using internet searches . The presentation is non-specific to the pathogen and many searching behaviors that an ill person with influenza might exhibit overlap with searching by individuals afflicted by other pathogens . Pathogens exhibiting a distinct clinical presentation described by disease-specific terms that are widely used by the general population might exhibit the clearest correlation of search terms with disease incidence . Additionally , prediction of incidence is more important for pathogens that exhibit strong temporal variation . Dengue exhibits both of these characteristics: It exhibits a more distinct clinical presentation than influenza , giving rise to more disease-specific terms; and , dengue incidence in many locations exhibits large interannual variability with incidence varying by as much as a factor of ten from one year to the next [9] . The dengue virus is an arboviral illness of the family Flaviviridae and consists of 4 antigenically distinct serotypes . It is transmitted by the bite of an infected mosquito ( Aedes aegypti poses the biggest threat to humans ) and infection by one serotype does not confer life-long immunity to another serotype . The incubation period is typically 4–7 days and may present with undifferentiated fever , petechiae rash , nuchal headache , myalgia and arthralgia . The severe clinical manifestation , dengue hemorrhagic fever ( DHF ) , is strongly associated with second infections and arises in around 3% of cases [10] . Prediction of outbreaks of dengue virus in countries with underdeveloped surveillance is of great importance to ministries of health and other public health decision makers who are often constrained by budget or man-power . The clinical presentation of dengue , although overlapping with other pathogens , is more specific to dengue than ILI is to influenza infection , and many of the search terms that individuals might search for when seeking information on dengue are specific to dengue ( as opposed to terms such as ‘cold’ ) . Thus , internet searches might exhibit stronger correlation with dengue incidence than influenza . Accurate predictions of dengue incidence might allow for more effective targeting of control measures such as vector control and preparation for surges in patients among hospitals , and may significantly increase the rapidity of dengue predictions in areas with less developed surveillance systems . In Thailand , dengue has been a significant source of morbidity and mortality for over 70 years . DHF was first observed in Bangkok in 1949 . The Thai Ministry of Public Health has conducted dengue surveillance since 1968 . Incidence in Bangkok varies widely from year from 15 , 000 cases to over 175 , 000 cases annually . In Singapore , DHF was a significant cause of childhood mortality , in the 1960s , 1970s and 1980s , prompting vector control efforts that reduced the density of Aedes mosquito breeding sites and precipitated a decline in the incidence of DHF in the late 1980s and early 1990s [11] . However , from the late 1990s onwards , there has been a resurgence of dengue fever despite low levels of Aedes mosquito breeding , culminating in the largest observed epidemic in Singaporean history of over 14 , 000 cases in 2005 . Although there is a trend towards an increase during the middle of the year , there is a wide variation of weekly incidence ranging from 32 to 713 cases during the 2004–2011 period . The ability to accurately predict a rise in incidence would be a useful way to trigger a series of clinical interventions ( deployment of medical teams , clearing of hospitial beds ) , and public health interventions ( escalation in surveillance , public health education and pre-emptive source reduction measures ) to reduce the transmission of dengue fever . Internet search term based surveillance could decrease delays associated with traditional surveillance systems and support under-developed systems .
Search data were downloaded from Google Insights for Search ( http://www . google . com/insights/search/ ) on February 18th , 2011 for Singapore and March 2nd , 2011 for Bangkok . Relevant search terms for both were selected by brainstorming common words used in searching for dengue . We searched terms that include words in all three of the official languages in Singapore; English , Chinese Malay and Tamil . Terms for both Singapore and Bangkok were classified into 3 categories: nomenclature , signs/symptoms and treatment . The search terms for the “full models” are shown in Figure 1 . Google Insights for Search provides related searches that generate a significant volume of results . All relevant related search data were retrieved . Google Insights for Search ignores capitalization , but treats misspellings and different orderings ( for example “symptoms flu” and “flu symptoms” ) as distinct searches . However , the volume of search data for these are small and none were included in model testing . Often , the Google Insight engine would only return data aggregated by month , because of uncertainty in weekly estimates in terms with low levels of search . For these terms a cubic spline was used to disaggregate the data to weekly responses ( using R's spline ( ) ) ; negative values resulting from the spline were set to 0 . The data were also regressed with the same model terms using the monthly aggregated data , and similar results were obtained ( see below ) . Importantly , Google Insight returns a sample of the actual search volume , so exact replication of the estimates of the model covariates is impossible . To correct for seasonal variation and confounding by time we included both the month of the year ( coded numerically as 1 for January , 2 for February , etc ) and a numeric code indicating week and year of the current data point ( given in R as the number of days since January 1st , 1970 ) . Epidemiologic surveillance data were obtained from the Singapore Ministry of Health website which conducts routine epidemiological data collection via the government polyclinics , public hospitals , clinical laboratories as well as via mandatory communicable disease reporting procedures [11] . Clinical and laboratory confirmed dengue fever cases have been reported to the Ministry of Health since 1977 and the data are aggregated by week . Thai monthly incidence data were gathered from the Thai Bureau of Epidemiology website . Since Google provides data on internet searches only since 2004 , we only considered dengue incidence data from 2004 . Incidence data for both Singapore and Bangkok are presented as the black lines in Figure 2 . We considered two outcomes , incident dengue cases and a binary outcome defined to be 1 during periods of high incidence and 0 otherwise . Multiple linear regression , negative binomial regression and generalized boosted regression ( GBR ) were used to model the weekly incidence of dengue fever using internet search terms [12] . A backwards and forwards step procedure was used to find the linear regression model that maximizes the Akaike information criterion ( AIC ) . Negative binomial regression fit with the full set of search terms in each location was chosen over Poisson regression due to over-dispersion of the search term data . GBR models were fit using the gbm package in R [13] . Candidate models were trained using 2005–2010 data and used to predict 2011 incidence . Inclusion of 2004 data from Singapore reduced the predictive accuracy of the model . Because the predictions were not qualitatively different , and included a nearly overlapping set of covariates when 2004 was or was not included , we chose to optimize our predictions of incidence in later years by dropping 2004 from the models . To choose between the multiple linear regression , negative binomial regression and GBR models , we determined the model with the largest correlation between the 2010 prediction and lagged incidence . This model was then cross-validated to evaluate prediction performance . We used leave one out cross-validation and an expanding prediction window ( both weekly and yearly , forward and backward ) for search data dated between January 2005 and through December 2010 , and evaluated the normalized root mean square error ( NRMSE ) of the predicted values of the left-out data from the observed incidence . In addition to models predicting incidence , logistic regression and Support Vector Machine ( SVM ) models were used to predict periods of high incidence [12] . We built models for three different high incidence thresholds defined as the 50th , 75th and 90th percentiles of numbers of cases over the period 2005–2011 . Model performance was assessed using the area under the receiver operating characteristic curve ( AUC ) for leave-one-out prediction . All statistical analyses were conducted in R version 2 . 12 . 2 ( R Core Development Team ) .
The AIC step-down model outperformed the GBR and negative binomial model for predicting numbers of incident cases and was chosen as optimal in both Singapore and Bangkok . The best fitting AIC step-down models have the predictor search terms shown in Figure 1 . Table 1 shows the model diagnostics comparing the step-down and full models for Singapore and Bangkok . A multiple time series plot showing normalized dengue incidence , the results of the optimized model fits and the error between predicted and observed incidence is presented as Figure 2 . To assess the performance of the prediction on data that was not used to fit the model , we used multiple cross-validation techniques . We predicted incidence in 2010 in both locations using models fit to data from 2005–2010 . Correlation between predictions of dengue in 2010 and observed dengue incidence for both Singapore and Bangkok are reported in Table 1 . We also assessed predictions for single and multiple observations that were left out of the data set used to fit the model . These results ( reported in the Supporting Information S1 ) indicate a good fit of the step-down model relative to the full model . Additionally , the prediction errors are low in the leave-one-week-out case and the leave-52-weeks-out case . We also see poor performance of the negative binomial model relative to the other models . For both Singapore and Bangkok , logistic regressions and SVM models were fit to predict the binary outcome of incidence above or below a threshold . Figure 3 summarizes the prediction of the SVM model in Singapore ( a similar graph presenting the Bangkok SVM model is presented in the Supporting Information S1 ) , and Table 2 presents the AUC and optimal sensitivities and specificities for the logistic and SVM models for each of the three cutoffs . We can see good prediction for the median and 75th percentile cutoffs . We compared the performance of our model to a lag-1 autoregressive model using only dengue surveillance data from the last week ( Singapore ) or month ( Bangkok ) to predict the next observation . In Singapore , this model performs well , yielding a correlation between predictions and observed cases of 0 . 950 . In Bangkok , the model performs much worse than models using search terms with a correlation of 0 . 766 ( for comparison , the 8 search term model above has a correlation of 0 . 943 ) . However , delays in compilation of these reports , especially in other locations could mean that these data would be unavailable for an autoregressive prediction model .
We have found that specific internet search terms are highly correlated with dengue incidence . Our best model for data from Singapore which included 16 terms showed a correlation of 0 . 931 with observed dengue incidence and an . The 8 term model for Bangkok performs equally well with a correlation of 0 . 869 and an . Out-of-sample predictions are predictably lower , but not significantly so . Our predictions of time periods with high dengue incidence are very accurate with sensitivities and specificities of 0 . 861–1 . 00 and 0 . 765–1 . 00 for multiple thresholds in each location . Together , these results demonstrate the viability of this data stream in supporting dengue surveillance . Our model performed similarly to models built in other efforts to predict influenza incidence using internet search terms . Ginsberg et al . found a correlation of 0 . 90 for influenza incidence in the US using a model that included 45 search terms [1] , and Polgreen et al . fit a series of models to influenza data in the United States and all had values of [4] . In out-of-sample prediction , our models performed slightly worse than the models of influenza produced by Ginsberg et al , which found a correlation of 0 . 97 ( compared to 0 . 921 in Bangkok and 0 . 785 in Singapore ) . It should be noted that our model produces predictions for the entire year including high and low incidence seasons , whereas the models of Ginsberg produce predictions for only the influenza season . The accuracy of our predictions may be due to the clear clinical presentation of severe dengue . The larger interannual variability may also allow us to disentangle seasonal search behavior from dengue specific search behavior . The search terms included in the models include nomenclature terms , terms describing signs and symptoms as well as treatment seeking . Interestingly , 11 of the 13 search terms that were found to be significant in our final model for Singapore were in English . This suggests that the typical language used for health seeking behavior in Singapore is English . In Bangkok , we also found that three of the seven significant terms are English . We validated the candidate models using leave-one-observation-out , leave-one-year-out and forward and backward validation techniques . The model performance was fairly consistent across these different approaches . In our validations , we found one year with large incidence to be highly influential for the performance of our model ( see Supporting Information S1 ) . We expect that including future years with large incidence might further improve our results . Singapore has an extremely well developed dengue surveillance system that makes reported cases available to policy makers and the general public with a delay of around one week . In a setting with this rapidity of reporting , it is challenging for an internet search term model to return results more quickly and with better performance than a model that uses only reported cases to predict future cases [14] . This point has been demonstrated elsewhere for predicting consumer behavior: predictive search term-based models perform better when used in conjunction with rich independent data sets [15] . Thus , in Singapore , this tool might best be used as a supplement to existing surveillance systems . However , in other settings , with less developed surveillance systems , an internet search term-based system may yield significant gains in the rapidity of predictions . In Thailand , there are significant delays in the reporting of cases from many areas of the country . Our model may give significant improvements in settings with significant delays . It is conceivable that some dengue-endemic settings in South and Southeast Asia may have significant internet use before surveillance systems are developed and thus an internet search term-based model may be a proxy for routine surveillance in these settings . Caution must be used when generalizing our method to other settings . Even though we have chosen two settings that have very different rates of internet usage , both countries are of higher income than many of the countries in the region . However , it is reasonable to assume increasing internet penetration in the future . Individual models need to be developed for specific settings using local surveillance data and search terms . This effort shows that this approach may have promise in other settings . There are several other limitations to our work . Internet searching behavior is susceptible to the impact of media reports as has been found for influenza systems [1] , [16] . The rate of internet use and the rate of health information seeking in this setting may be changing over time and thus our parameters might need to shift over time to incorporate the impact of these changes . Although not affecting performance here , future outbreaks of other clinically similar diseases such as chikungunya may challenge the performance of our model for dengue . Finally , the Google Insight tool returns a sample of actual search data and limits the availability of search terms for which there are very few returns , often aggregating these terms to a large temporal discretization . This limits the utility of these terms for the purposes of prediction . Search query surveillance is rapidly expanding into many areas of public health including the surveillance of noninfectious diseases and to influencing policy domains [17]–[21] . The current work demonstrates the utility of using search query surveillance to forecast the incidence of a tropical infectious disease . Additionally , and importantly , we have constructed forecasting models using freely available search query data from Google Insights and publicly available surveillance data from Singapore and Bangkok . In addition , we have developed these models using open source software from the R statistical project . Our approach can be readily adapted to other settings where other proprietary efforts can not be implemented . The approach may be an important tool in many dengue endemic settings in supporting the public health response to dengue . | Improvements in surveillance , prediction of outbreaks and the monitoring of the epidemiology of dengue virus in countries with underdeveloped surveillance systems are of great importance to ministries of health and other public health decision makers who are often constrained by budget or man-power . Google Flu Trends has proven successful in providing an early warning system for outbreaks of influenza weeks before case data are reported . We believe that there is greater potential for this technique for dengue , as the incidence of this pathogen can vary by a factor of ten in some settings , making prediction all the more important in public health planning . In this paper , we demonstrate the utility of Google search terms in predicting dengue incidence in Singapore and Bangkok , Thailand using several regression techniques . Incidence data were provided by the Singapore Ministry of Health and the Thailand Bureau of Epidemiology . We find our models predict incident cases well ( correlation greater than 0 . 8 ) and periods of high incidence equally well ( AUC greater than 0 . 95 ) . All data and analysis code used in our study are available free online and can be adapted to other settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"mathematics",
"epidemiology",
"statistics",
"infectious",
"disease",
"epidemiology",
"global",
"health",
"dengue",
"viral",
"diseases",
"neglected",
"tropical",
"diseases",
"public",
"health",
"statistical",
"methods"
] | 2011 | Prediction of Dengue Incidence Using Search Query Surveillance |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.